INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND

INFORMATICS (ICI3’2K24)
TABLE OF CONTENTS

PAGE
SI.NO TITLE OF THE PAPER COLLEGE NAME
NUMBER
KAMARAJ COLLEGE OF
REAL TIME BUS LOCATOR
1 ENGINEERING AND 1
APPLICATION
TECHNOLOGY
AI INFUSED WRIST COMPANION FOR
KGISL INSTITUTE OF
MENTALLY RETARDED
2 TECHNOLOGY 2
HOSPITALISED INDIVIDUALS

KAMARAJ COLLEGE OF
ETHICAL CONSIDERATIONS IN
ENGINEERING AND
3 SOCIAL NETWORK ANALYSIS 3
TECHNOLOGY

FOOD WASTE MANAGEMENT: AN

KAMARAJ COLLEGE OF
IDEA OF CONNECTING ORPHANAGES
4 ENGINEERING AND 4
OR NEEDY PEOPLE WITH DONORS
TECHNOLOGY
USING A MOBILE APPLICATION
AI-ENABLED DISASTER
MANAGEMENT SYSTEM : ENABLING
KARPAGAM COLLEGE OF
5 EFFECTIVE COMMUNICATION 5
ENGINEERING
BETWEEN RESOURCE AGENCIES TO
SUPPORT PEOPLE DURING CRISIS
KONGU POLYTECHNIC COLLEGE
6 CYBER SECURITY 6
PERUNDURAI.
DESIGN OF AN ASYMMETRIC CPW KONGUNADU COLLEGE OF
7 FED WIDEBAND MIMO ANTENNA ENGINEERING AND 7
FOR 5G APPLICATION TECHNOLOGY
AN ECONOMICAL OFF SEASON FARM
EQUIPMENT RENTAL PLATFORM KARPAGAM COLLEGE OF
8 8
USING REACT AND DJANGO FRAME ENGINEERING
WORK
MULTILINGUAL LEGAL AI KGISL INSTITUTE OF
9 9
ASSISTANT TECHNOLOGY
HEALTH MONITORING SYSTEM FOR M KUMARASWAMY COLLEGE OF
10 10
TYPE-2 DIABETES ENGINEERING
AGRI INTELLECT EMPOWERING
ANJALAI AMMAL MAHALINGAM
11 FARMERS WITH SMART 11
ENGINEERING COLLEGE
AGRICULTURAL SOLUTION
DESIGN AND IMPLEMENTATION OF
M. KUMARASAMY COLLEGE OF
12 SECURITY BASED PARKING SYSTEM 12
ENGINEERING, KARUR
FOR SMART CITIES
M KUMARASAMY COLLEGE OF
13 CAR CLUB RENTAL 13
ENGINEERING

INTELLECTUAL AGRIBOT FOR KGISL INSTITUTE OF

14 14
BANANA FARMING TECGNOLOGY
 INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND
INFORMATICS (ICI3’2K24)
TABLE OF CONTENTS

DETECTION OF TIME-STAMPED M.A.M.COLLEGE OF

15 15
SUSPICIOUS ACTIVITY IN A ENGINEERING AND
SURVEILLANCE CAMERA FOOTAGE TECHNOLOGY
ADVANCED REAL TIME OBJECT M.A.M COLLEGE OF
16 DETECTION BASED ON EDGE GROUP ENGINEERING AND 16
FOR SMART VIDEO TECHNOLOGY
MULTI FACTOR AUTHENTICATION KAMARAJ COLLEGE OF
17 FOR WEB APPLICATION USING MAC ENGINEERING AND 17
ADDRESS TECHNOLOGY
KAMARAJ COLLEGE OF
18 PSYCHE PLAYER ENGINEERING AND 18
TECHNOLOGY
NON INVASIVE FETAL MONITORING PSNA COLLEGE OF
19 WITH COMBINED PCG- PULSE ENGINEERING AND 19
MONITORING SYSTEM TECHNOLOGY
COMPARATIVE ANALYSIS OF 3D
20 PSNACET 20
ROOT CANAL MODELS
SUSTAINABLE TOURISM M KUMARASAMY COLLEGE OF
21 21
DEVELOPMENT SYSTEM ENGINEERING
ONLINE RESIDENTIAL RENTAL M. KUMARASAMY COLLEGE OF
22 22
MANAGEMENT ZONE ENGINEERING
INTEGRATING LANGUAGE
TRANSLATOR TO ENHANCE KARPAGAM COLLEGE OF
23 24
SECURITY AND ACCESS CONTROL ENGINEERING
USING API
SRI CHANDRASEKHARENDRA
SARASWATHI VISWA
24 EASY BUY 25
VIDYALAYAISWA
MAHAVIDYALA
A SMART SYSTEM FOR ACCIDENT
RAMCO INSTITUTE OF
25 AVOIDANCE USING RAPID EYE 26
TECHNOLOGY
MOTION DETECTION
DETECTION OF FETAL SCAD COLLEGE OF
26 ANENCEPHALY IN BRAIN BASED ON ENGINEERING AND 27
ENSEMBLES LEARNING ENGINEERING
A NOVEL PARADIGM OF INDOOR
NAVIGATION SYSTEM USING LI-FI
KNOWLEDGE INSTITUTE OF
27 TECHNOLOGY 28
TECHNOLOGY

AUTOMATIC ROOM LIGHTS

SATHYABAMA INSTITUTE OF
28 CONTROL BY USING ARDUINO 29
SCIENCE AND TECHNOLOGY
DESIGN
LEGALEASE CHATBOT - BRIDGING
ANJALAI AMMAL MAHALINGAM
29 LEGAL KNOWLEDGE GAPS FOR 30
ENGINEERING COLLEGE
MARGINALIZED COMMUNITIES
 INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND
INFORMATICS (ICI3’2K24)
TABLE OF CONTENTS

THREATS ANALYSIS AND

DETECTION FOR DARKWEB
30 JP COLLEGE OF ENGINEERING 31
DATASET USING MACHINE
LEARNING APPROACHES
TOXIC COMMENT CLASSIFICATION PAAVAI ENGINEERING COLLEGE
31 33
USING MACHINE LEARNING (AUTONOMOUS)
SPECTRAL SUBTRACTION
ALGORITHM FOR NOISE
32 SUPPRESSION IN AUDIO SINGALS : A JP COLLEGE OF ENGINEERING 34
COMPARATIVE ANALYSIS WITH
KRISP ALGORITHM
FREEBEEZ:A FREELANCERS M.KUMARASAMY COLLEGE OF
33 35
COMMUNITY WEBSITE ENGINEERING
AUTOMATED SMOKE AND GAS
34 LEAKAGE DETECTOR WITH IOT JP COLLEGE OF ENGINEERING 36
MONITORING SYSTEM
BLOOD BANK MANAGEMENT M KUMARASAMY COLLAGE OF
35 37
SYSTEM ENGINEERING
ACCURATE OVARIAN CYST
CLASSIFICATION WITH A
36 PAAVAI ENGINEERING COLLEGE 38
LIGHTWEIGHT DEEP LEARNING
MODEL FOR ULTRASOUND IMAGES.
37 IOT BASED SMART NEBULIZER P.S.R ENGINEERING COLLEGE 39
LEVERAGING YOLO NAS AND
CONTRASTIVE LEARNING FOR NEXT- KARPAGAM COLLEGE OF
38 40
GENERATION RETAIL ENVIRONMENT ENGINEERING
ANALYSIS
ALUMNI WEB PORTAL EMERGED KGISL INSTITUTE OF
39 42
WITH DEEP LEARNING TECHNOLOGY
FRICTION STIR WELDING OF AA2219
PERIYAR MANIAMMAI
ALUMINIUM ALLOY WITH THREE
40 INSTITUTE OF SCIENCE AND 43
DIFFERENT PROFILE TOOLS AND
TECHNOLOGY
THEIR CHARACTERIZATION
THE HYBRID MODEL FOR OPTIMIZED
EXTREME HEATWAVE PREDICTION
KAMARAJ COLLEGE OF
USING ADAPTIVE NEURO-FUZZY
41 ENGINEERING AND 44
INFERENCE SYSTEMS
TECHNOLOGY

DESIGN AND ANALYSIS OF A EQUAL

DR.MAHALINGAM COLLEGE OF
SPLIT COMPACT WILKINSON POWER
42 ENGINEERING AND 45
DIVIDER FOR WIRELESS
TECHNOLOGY
COMMUNICATION SYSTEMS
SATHYABAMA INSTITUTE OF
ADVANCED ADAPTIVE CRUISE
43 SCIENCE AND TECHNOLOGY , 46
CONTROL SYSTEM IN AUTONOMOUS
CHENNAI
VEHICLE
 INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND
INFORMATICS (ICI3’2K24)
TABLE OF CONTENTS

FABRICATION AND MECHANICAL

BEHAVIOR OF PERIYAR MANIAMMAI
44 AL20CO20FE20NI20MN20 HIGH INSTITUTE OF SCIENCE AND 47
ENTROPY ALLOY THROUGH TECHNOLOGY
MECHANICAL ALLOYING
SUSTAINING TRADITION: CHICKEN
45 RANCH ORGANIC E-COMMERCE JP COLLEGE OF ENGINEERING 48
EVOLUTION
AGRI-KET LIVE STOCK M. KUMARASAMAY COLLEGE
46 49
MANAGEMENT WEBPAGE OF ENGINEERING
IOT FARM BASED PROTECTION
47 P.S.R ENGINEERING COLLEGE 50
SYSTEM
DR MAHALINGAM COLLEGE OF
48 EXAMINATION HALL ALLOCATE ENGINEERING AND 51
TECHNOLOGY
COVER TIMING CHANNEL USING NANDHA ENGINEERING
49 52
IMAGE PROCESSING COLLEGE
PSNA COLLEGE OF
CHATBOT FOR GOVERNMENT
50 ENGINEERING AND 53
SCHEMES
TECHNOLOGY
PROSTHETIC SOLUTION FOR LOWER
51 LIMB AMPUTEES WITH ANKLE P.S.R ENGINEERING COLLEGE 54
MOVEMENT
AI ENHANCED RAILWAY PLATFORM
52 JP COLLEGE OF ENGINEERING 55
NAVIGATION FOR BLIND PEOPLE
FOODEVEN - FOOD ORDERING PSNA COLLEGE OF
53 SYSTEM USING MOBILE ENGINEERING AND 56
APPLICATION TECHNOLOGY
PPG INSTITUTE OF
54 FOOD WASTAGE PREVENTION APP 57
TECHNOLOGY
M.A.M COLLEGE OF
55 ONLINE RATION SHOPPING ENGINEERING AND 58
PLATFORM TECHNOLOGY
HEART RESCUE: IOT INTEGRATION
VELALAR COLLEGE OF
FOR ADVANCED CPR FEEDBACK
56 ENGINEERING AND 59
AND DEFIBRILLATION IN CARDIAC
TECHNOLOGY
EMERGENCIES
SMART VEHICLE SAFETY SYSTEM
57 P.S.R ENGINEERING COLLEGE 60
FOR TWO WHEELER
REALTIME ROBOTIC AVATAR USING
FUZZY GAZE CLASSIFICATION SREE SAKTHI ENGINEERING
58 61
ALGORITHM FOR PEOPLE WITH COLLEGE KARAMADAI
DISABILTIY
DESIGN AND ANALYSIS OF
ELLIPTICAL AND CIRCULAR RAMCO INSTITUTE OF
59 62
ANTENNAS FOR UWB APPLICATIONS TECHNOLOGY
 INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND
INFORMATICS (ICI3’2K24)
TABLE OF CONTENTS

AN EXPERIMENTAL STUDY ON
P.A. COLLEGE OF ENGINEERING
60 MECHANICAL PROPERTIES OF 63
AND TECHNOLOGY
COUROUPITA GUIANENSIS TREE
TAGORE INSTITUTE OF
FLOOD MANAGEMENT AND EARLY
61 ENGINEERING AND 64
MONITORING SYSTEM
TECHNOLOGY
M.A.M COLLEGE OF
62 IOT BASED INTELLIGENCE ROBOT ENGINEERING AND 65
TECHNOLOGY
AUGMENTED REALITY WITH
SREE SAKTHI ENGINEERING
62 DYNAMIC RENDERING FOR 66
COLLEGE
EDUCATION
IDENTIFICATION OF LIVER HEALTH
RAMCO INSTITUTE OF
63 USING MACHINE LEARNING 67
TECHNOLOGY
ALGORITHMS
KAMARAJ COLLEGE OF
AMAZON RECOMMENDATION
64 ENGINEERING AND 68
SYSTEM
TECHNOLOGY
A ROBOTIC STEWARD FOR MARINE VELALAR COLLEGE OF
65 CLEANUP AND ECO-SYSTEM ENGINEERING AND 69
SURVEILLANCE TECHNOLOGY
M.KUMARASAMY COLLEGE OF
66 ONLINE LOCAL FARMER MARKET 70
ENGINEERING
VEHICLE POWER GENERATION
NPR COLLEGE OF ENGINEERING
67 SYSTEM FOR STREET LIGHTING AND 71
AND TECHNOLOGY
VEHICLE POWER STATION
OPTIMIZING DATA MINING
RESEARCH: LEVERAGING WEKA'S
68 T. JOHN COLLEGE 72
BEST FIT TOOL FOR ENHANCED
EFFECTIVENESS
WATER RESOURCES MONITING
69 SYSTEM USING MACHINE LEARNING JP COLLEGE OF ENGINEERING 73
APPROACH
M.A.M. COLLEGE OF
ADVANCED AUTOMATED BANKING
70 ENGINEERING AND 74
TERMINAL
TECHNOLOGY
PHISHING MAIL DETECTION USING KATHIR COLLEGE OF
71 76
LARGE LANGUAGE MODELS ENGINEERING
EMERGENCY CONNECT: ANDROID SRI RAMAKRISHNA
72 77
APP ENGINEERING COLLEGE
SYNTHESIS OF ANATASE GRADE
TIO2 BY OXIDATION CUM LEACHING KALASALINGAM ADAMIC
73 78
PROCESS FROM TITANIUM BEARINGS RESEARCH AND EDUCATION
MINERALS
ENHANCING MICROBIAL FUEL CELL
KALASALINGAM ACADEMY OF
74 EFFICIENCY BY USING EGGSHELL 79
RESEARCH AND EDUCATION
POWDER WITH SACCHAROMYCES
 INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND
INFORMATICS (ICI3’2K24)
TABLE OF CONTENTS

CEREVISIAE IN BATCH PROCESSES

DETECTING AND CLASSIFYING NOORUL ISLAM CENTRE FOR
75 MALWARE WITH MACHINE HIGHER EDUCATION, 80
LEARNING KUMARACOIL
DR.MAHALINGAM COLLEGE OF
ALEXA HOME AUTOMATION SYSTEM
76 ENGINEERING AND 81
USING NODEMCU ESP8266
TECHNOLOGY, POLLACHI
DR. MAHALINGHAM COLLEGE
77 MODELLING OF EV USING SIMULINK OF ENGINEERING AND 82
TECHNOLOGY
A SMART OBSTACLE DETECTOR AND T.S SRINIVASAN CENTRE FOR
78 PERSONAL VOICE ASSISTANT FOR POLYTECHNIC COLLEGE AND 83
BLIND PEOPLE USING IOT ADVANCED TRAINING
HEALTHCARE SYSTEM USING A.V.C COLLEGE OF
79 84
EMOTIONAL INTELLIGENCE ENGINEERING
MICROWAVE SENSING OF QUALITY TAGORE INSTITUTE OF
80 ATTRIBUTES OF AGRICULTURAL ENGINEERING AND 85
AND FOOD PRODUCTS TECHNOLOGY
DETECTION OF CORONA VIRUS FRANCIS XAVIER ENGINEERING
81 USING LUNG X-RAY IMAGES USING COLLEGE, MEPCO SCLENK 86
MACHINE LEARNING ENGINEERING COLLEGE,
M.KUMARASAMY COLLEGE OF
82 87
LIBRARIAN’S DELIGHT ENGINEERING
NON-INVASIVE BLOOD PRESSURE
MONITORING USING
SRI RAMAKRISHNA
83 PHOTOPLETHYSMOGRAPHY 88
ENGINEERING COLLEGE
SIGNALS: A COMPREHENSIVE
INVESTIGATION
INTRUSION DETECTION WHILE
KATHIR COLLEGE OF
84 ACCESSING LOG DATA FILES IN 89
ENGINEERING
CLOUD DATALAKE ARCHITECTURE
SUSTAINABLE DEVELOPMENT
THROUGH POVERTY ANALYSIS SRM INSTITUTE OF SCIENCE
85 90
USING RANDOM FOREST AND TECHNOLOGY
ALGORITHM
SREE SAKTHI ENGINEERING
86 SMART KITCHEN USING IOT 91
COLLEGE
VELALAR COLLEGE OF
AN IOT BASED MEDI-KIT FOR
87 ENGINEERING AND 92
PASSENGERS IN TRAIN
TECHNOLOGY
OPTICAL SPECTROSCOPY METHOD
DHIRAJLAL GANDHI COLLEGE
88 OF DETERMINING AND ANALYSING 93
OF TECHNOLOGY
BLOOD GLUCOSE LEVEL IN HUMAN
DETECTION OF OVARIAN CANCER
KGISL INSTITUTE OF
89 BASED ON CLINICAL DATA USING 94
TECHNOLOGY
MACHINE LEARNING APPROACHES
 INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND
INFORMATICS (ICI3’2K24)
TABLE OF CONTENTS

EXPLORING REACT-NATIVE WITH

M. KUMARASAMY COLLEGE OF
90 NLP: A CASE STUDY IN CONTENT 95
ENGINEERING
RECOMMENDATION SYSTEMS
DR.MAHALINGAM COLLEGE OF
A DYNAMIC POWER EFFICIENT CMOS
91 ENGINEERING AND 96
COMPARATOR
TECHNOLOGY
CULTIVATING AGRICULTURAL
INSIGHTS: HARNESSING ERODE SENGUNTHAR
92 97
HYPERSPECTRAL DATA FOR BIG ENGINEERING COLLEGE
DATA ANALYTICS
VISUAL FIRE ALARM SYSTEM:
NANDHA ENGINEERING
93 LEVERAGING OPENCV FOR 98
COLLEGE
DETECTION
BLOCK CHAIN WASTE
NANDHA ENGINEERING
94 MANAGEMENT USING SECURE DATA 99
COLLEGE - ERODE
STANDARD
BIOELECTRICITY GENERATION
FROM BANANA PEELS WASTE USING
95 KALASALINGAM UNIVERSITY 100
DOUBLE CHAMBER MICROBIAL
FUELCELL.
MINDMATE CHATBOT: AN AI-DRIVEN
KGISL INSTITUTE OF
96 APPROACH TO MENTAL HEALTH 101
TECHNOLOGY
CARE
DYNAMIC SCHEDULING IN EDGE-
CLOUD COMPUTING ENVIRONMENTS KONGUNADU ARTS AND
97 102
USING SEAGULL OPTIMISATION SCIENCE COLLEGE
ALGORITHM
LEAF DISEASE DETECTION USING KATHIR COLLEGE OF
98 103
DEEP LEARNING ENGINEERING
GANVIZ:TEXT TO IMAGE
ERODE SENGUNTHAR
99 GENERATION USING GENERATIVE 104
ENGINEERING COLLEGE
ADVERSIAL NETWORK
FORECASTING VARIOUS CHILDBIRTH
NANDHA ENGINEERING
100 METHODS USING MACHINE 105
COLLEGE
LEARNING
DETECTION OF VIRTUAL PRIVATE AVINASHILINGAM INSTITUTE
NETWORKS ENCRYPTED TRAFFIC FOR HOME SCIENCE AND
101 106
USING MACHINE AND ENSEMBLE HIGHER EDUCATION FOR
LEARNING ALGORITHMS WOMEN
UNLOCKING THE POTENTIAL OF M KUMARASAMY COLLEGE OF
102 107
INDUSTRIAL AUTOMATION ENGINEERING
ANGEL COLLEGE OF
ONLINE BLOOD BANK USING
103 ENGINEERING AND 108
CHATBOT
TECHNOLOGY
FOOD CALORIES DETECTION USING
ERODE SENGUNTHAR
104 GEMINI API 109
ENGINEERING COLLEGE
 INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND
INFORMATICS (ICI3’2K24)
TABLE OF CONTENTS

ANGEL COLLEGE OF
SMART TRAVEL PLANNER WEBSITE
105 ENGINEERING AND 110
FOR TOURIST USING JAVA SCRIPT
TECHNOLOGY
VELS INSTITUTE OF SCIENCE
WINDOWS 14: UI AND UX DESIGN
106 TECHNOLOGY AND ADVANCED 111
INNOVATION
STUDIES
DESIGN AND ANALYSIS OF
K. RAMAKRISHNAN COLLEGE OF
107 BANDPASA FILTER IN C BAND 112
TECHNOLOGY
APPLICATION.
IoT ENABLED NON-INVASIVE
BIO IMPEDANCE SPECTROSCOPY SRI RAMAKRISHNA
108 113
BASED HUMAN HYDRATION ENGINEERING COLLEGE
MONITOR FOR DIALYSIS PATIENTS
CARE CONNECT FOR ORPHANAGE SATHYABAMA INSTITUTE OF
109 114
CHILDREN SCIENCE AND TECHNOLOGY
VELALAR COLLEGE OF
TRANSPARENT ELECTRONIC VOTING
110 ENGINEERING AND 115
USING BLOCKCHAIN
TECHNOLOGY
AN ANOMALY DETECTION BASED ON
BANNARI AMMAN INSTITUTE OF
111 MACHINE LEARNING IN WIRELESS 116
TECHNOLOGY
NETWORKS
DEEP LEARNING DRIVEN BEARD
K.RAMAKRISHNAN COLLEGE OF
112 RECOGNITION SYSTEM WITH 117
TECHNOLOGY
AUTOMATED ALERTS
SMART PLANT MONITORING SYSTEM K RAMAKRISHNAN COLLEGE OF
113 118
USING IOT TECHNOLOGY
EFFICIENT ERROR DETECTION
K RAMAKRISHNAN COLLEGE OF
114 SCHEMES FOR ECSM WINDOW 119
TECHNOOLGY
METHOD
NANDHA ENGINEERING
SECURITY AND PRIVACY FOR
115 COLLEGE, AUTONOMOUS - 120
HEALTHCARE BLOCKCHAIN
ERODE
PLANT DISEASE CLASSIFICATION K. RAMAKRISHNAN COLLEGE OF
116 121
FROM LEAF IMAGES USING SVM TECHNOLOGY
DESIGN AND IMPLEMENTATION OF AVINASHILINGAM INSTITUTE OF
117 SMART HOME AUTOMATION SYSTEM HOME SCIENCE AND HIGHER 122
USING RASPBERRY PI EDUCATION FOR WOMEN

AUTOMATED WATER SURFACE KGISL INSTITUTE OF

118 123
CLEANING SYSTEM USING ESP8266 TECHNOLOGY

MANUFACTURING OF AMMONIUM
119 KALASALINGAM ACADEMY OF 124
SULPHATE BY DIRECT
RESEARCH AND EDUCATION
NEUTRALISATION
DETECTION OF HEAVY METAL IONS SRI RAMAKRISHNA
120 125
USING ELECTROCHEMICAL PROCESS ENGINEERING COLLEGE
 INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND
INFORMATICS (ICI3’2K24)
TABLE OF CONTENTS

AVINASHILINGAM INSTITUTE
HUMAN EYE BALL GAZING AND
FOR HOME SCIENCE AND
121 FACE BASED ULTRA DIGITAL 126
HIGHER EDUCATION FOR
SECURITY SYSTEM USING AI
WOMEN
DETECTING LUNG CANCER IN CT
NANDHA ENGINEERING
122 SCAN USING CNN A DEEP LEARNING 127
COLLEGE - ERODE
APPROACH.
AN EXPERIMENTAL STUDY ON
P.A.COLLEGE OF ENGINEERING
123 MECHANICAL PROPERTIES OF 128
AND TECHNOLOGY
BORASSUS FLABELLIFER
DETECTION OF FRAUDULENT JAI SHRIRAM ENGINEERING
124 129
ACTIONS IN CARD TRANSACTIONS COLLEGE, TIRUPUR
LUNG TUMOR SEGMENTATION AND PSNA COLLEGE OF
125 STAGE DETECTION USING CT SCAN ENGINEERING AND 130
IMAGES TECHNOLOGY
EMPOWERING TAMIL NADU:
PSNA COLLEGE OF
CENTRALIZED GRIEVANCE REDRESS
126 ENGINEERING AND 131
AND MONITORING SYSTEM
TECHNOLOGY
MODEL PREDICTIVE CURRENT DR.MAHALINGAM COLLEGE OF
127 CONTROL OF REDUCED SWITCH ENGINEERING AND 132
MULTILEVEL INVERTER TECHNOLOGY
DR.MAHALINGAM COLLEGE OF
PCB DESIGN OF H-BRIDGE USING
128 ENGINEERING AND 133
KICAD
TECHNOLOGY
VELSINSTITUTE OF
SUPPLY CHAIN MANAGEMENT ON
129 SCIENCE,TECHNOLOGY & 134
BLOCKCHAIN
ADVANCED STUDIES
DR.MAHALINGAM COLLEGE OF
ACCIDENT DETECTION AND ALERT
130 ENGINEERING AND 135
SYSTEM
TECHNOLOGY
AREA EFFICIENT MODIFIED
K RAMAKRISHNAN COLLEGE OF
131 COMPRESSOR BLOCKS FOR IMAGE 136
TECHNOLOGY
MULTIPLICATION
AUTOMATIC TEMPERATURE DR. MAHALINGAM COLLEGE OF
132 CONTROL SYSTEM IN POULTRY ENGNERRING AND 137
FARM TECHNOLOGY
LEVERAGING SENTIMENTAL
DR.MAHALINGAM COLLEGE OF
ANALYSIS FOR E-COMMERCE
133 ENGINEERING AND 138
INSIGHTS
TECHNOLOGY

SRI RAMAKRISHNA
STANDALONE NON-INVASIVE
134 ENGINEERING COLLEGE 139
GLUCOSE AND CHOLESTEROL
COIMBATORE
MONITOR
BRAIN TUMOR DETECTION USING ML
135 VELS UNIVERSITY 140
ALGORITHMS
 INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND
INFORMATICS (ICI3’2K24)
TABLE OF CONTENTS

DR MAHALINGAM COLLEGE OF
MALICIOUS MOBILE APPLICATION
136 ENGINEERING AND 141
DETECTION
TECHNOLOGY
DR.MAHALINGAM COLLEGE OF
IOT IN GESTURE - BASED CONTROL
137 ENGINEERING AND 142
FOR SMART HOMES
TECHNOLOGY
DR.MAHALINGAM COLLEGE OF
IOT IN RETAIL FOR SMART SHELF
138 ENGINEERING AND 143
MANAGEMENT
TECHNOLOGY
ACTIVE PROSTHETIC ANKLE AND
ADAPTIVE EQUIPMENT FOR BIKE VIVEKANANDHA COLLEGE OF
139 144
RIDING AND LOCOMOTION IN ENGINEERING FOR WOMEN
LOWER LIMB AMPUTEES
AUTOMATIC HAND GLOVES SRI RAMAKRISHNA
140 145
THERAPY SYSTEM ENGINEERING COLLEGE
AI-DRIVEN EXPLORATION OF
SRI RAMAKRISHNA
141 PATIENT RECORDS FOR IMPROVED 146
ENGINEERING COLLEGE
DIAGNOSIS.
DR.MAHALINGAM COLLEGE OF
DRIP INFUSION MONITORING
142 ENGINEERING AND 147
SYSTEM
TECHNOLOGY
BRAIN NEOPLASM IDENTIFICATION JAI SHRIRAM ENGINEERING
143 148
USING DEEP LEARNING COLLEGE
SECURE DIAGNOSIS: AUTOMATING
WIRESHARK FOR MEDICAL DATA SRI RAMA KRISHNA
144 149
ENCRYPTION ANALYSIS ENGINEERING COLLEGE

DESIGN AND DEVELOPMENT OF AN

SRI RAMAKRISHNA
145 AUTOMATIC WOUND DRESSING 150
ENGINEERING COLLEGE
APPLICATOR
A NOVEL DESIGN OF BLDC MOTOR PERIYAR MANIAMMAI
146 WITH FUZZY LOGIC CONTROL INSTITUTE OF SCIENCE AND 151
SYSTEM TECHNOLOGY
PREDICTION OF LIVER STIFFNESS
SRI RAMAKRISHNA
147 USING ELASTOGRAPHIC IMAGES 152
ENGINEERING COLLEGE
(FIBROSCAN)
AUTOMATED CLASSIFICATION OF SRI RAMAKRISHNA
148 153
TEXT BASED ON DISEASE ENGINEERING COLLEGE
SAFE GUARDING ANDROID ECO
SYSTEM WITH INTELLIGENT P.S.R. ENGINEERING COLLEGE
149 154
MALWARE DEFENSE SIVAKASI

TWITTER - BASED ELECTRONIC SRI RAMAKRISHNA

150 155
HEALTH RECORDS ENGINEERING COLLEGE
ENHANCING INDEPENDENCE
PAAVAI COLLEGE OF
151 VISUALLY IMPAIRED ASSISTANCE 156
ENGINEERING
USING YOLO TECHNOLOGY
 INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND
INFORMATICS (ICI3’2K24)
TABLE OF CONTENTS

PREDICTIVE CYBER ATTACK BY PAAVAI COLLEGE OF

152 157
USING XGBOOST ALGORITHM ENGINEERING
ENCRYPTED HONEY PASSWORD
PAAVAI COLLEGE OF
153 AUTHENTICATION FOR ONLINE 158
ENGINEERING
BANKING SYSTEM
DATA SECURE DEDUPLICATION AND
PAAVAI COLLEGE OF
RECOVERY BASED ON PUBLIC KEY
154 ENGINEERING 159
ENCRYPTION WITH KEYWORD
SEARCH
AGRICULTURE AND RURAL
DEVELOPMENT: EMPOWERING
KARPAGAM COLLEGE OF
155 COMMUNITIES FOR SUSTAINABLE 160
ENGINEERING
GROWTH

CT TRANSLATOR: A PERSONALIZED
KARPAGAM COLLEGE OF
156 TRANSLATION SOFTWARE 161
ENGINEERING
TOYS AND GAMES KARPAGAM COLLEGE OF
157 162
ENGINEERING
TEXCO:A COMPREHENSIVE DESKTOP
SOLUTION FOR REAL-TIME MARKET
KARPAGAM COLLEGE OF
158 INSIGHTS, ORDER TRACKING, AND 163
ENGINEERING
TRADING IN THE TEXTILE SECTOR

SMART TUTOR: PERSONALIZED

KARPAGAM COLLEGE OF
159 LEARNING REMEDIATION TOOL 164
ENGINEERING
A GEOGRAPHICAL INFORMATION
SYSTEM BASED APPROACH FOR KARPAGAM COLLEGE OF
160 165
EWASTE FACILITY LOCATOR ENGINEERING

REAL TIME BUS TRACKING SYSYTEM

KARPAGAM COLLEGE OF
161 : WITH LOCATION AND ARRIVAL , 166
ENGINEERING
DESTINATION TIME PREDICATIONS
AN AGRICULTURE PRODUCTS
KARPAGAM COLLEGE OF
162 EXPLORER SYSTEM THROUGH 167
ENGINEERING
ONLINE
AI-ENABLED DISASTER
MANAGEMENT SYSTEM : ENABLING
EFFECTIVE COMMUNICATION KARPAGAM COLLEGE OF
163 168
BETWEEN RESOURCE AGENCIES TO ENGINEERING
SUPPORT PEOPLE DURING CRISIS

ENHANCE THE QUALITY OF

EDUCATION IN RURAL AREAS IN KARPAGAM COLLEGE OF
165 169
SYSTEMATIC APPROACH ENGINEERING
 INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND
INFORMATICS (ICI3’2K24)
TABLE OF CONTENTS

DEVELOPMENT OFAN ENGLISH TO

HINDI LANGUAGE TRANSLATOR
KARPAGAM COLLEGE OF
166 TOOL FOR COMMUNICATION OF 170
ENGINEERING
GOVERNMENT ORGANIZATION

OPTIMIZING TICKET BOOKING

SYSTEMS IN THE TRAVEL AND
KARPAGAM COLLEGE OF
167 TOURISM INDUSTRY:A DATA DRIVEN 171
ENGINEERING
APPROACH

DISASTER MANAGEMENT OF IOT KARPAGAM COLLEGE OF

168 172
ENGINEERING
DIGITAL ASSISTANT FOR LEGAL
AWARENESS AND DESIGNING A KYR
KARPAGAM COLLEGE OF
169 KNOW-YOUR-RIGHTS FRAMEWORK 173
ENGINEERING
IN INDIA

AI ENABLED HRBOT HELPDESK FOR

KARPAGAM COLLEGE OF
170 EMPLOYEES COLLABORATION AND 174
ENGINEERING
COMMUNICATION
ENHANCING ACADEMIC INTEGRITY
THROUGH TECHNOLOGICAL
INNOVATIONS: A COMPREHENSIVE KARPAGAM COLLEGE OF
171 175
REVIEW OF PROCTORED ENGINEERING
EXAMINATION TOOLS AND
EMERGING TRENDS
ADVANCES IN REAL TIME
MONITORING OF AGRICULTURAL
KARPAGAM COLLEGE OF
172 DEVELOPMENT: ENHANCING 176
ENGINEERING
EFFICIENCY SUSTAINABILITY AND
RESILIENCE
ADVANCING DOMESTIC WASTE
MANAGEMENT THROUGH SMART
KARPAGAM COLLEGE OF
173 TECHNOLOGIES AND SUSTAINABLE 177
ENGINEERING
PRACTICES

VIRTUAL ZOO APPLICATION: LIVE

ANIMAL CAMS KARPAGAM COLLEGE OF
174 178
ENGINEERING

EMPOWERYOUNGRIGHTS:
DEVELOPMENT OF GAMIFIED
PLATFORM ON CHILDREN'S RIGHTS
KARPAGAM COLLEGE OF
175 TO INCREASE LEGAL LITERACY AND 179
ENGINEERING
AWARENESS AMONG CHILDREN IN
INDIA
 INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND
INFORMATICS (ICI3’2K24)
TABLE OF CONTENTS

COMMUNITY BASED REPORTING

AND MONITORING TOOL FOR
KARPAGAM COLLEGE OF
176 WOMEN’S SAFETY IN 180
ENGINEERING
COLLEGES/UNIVERSITIES

A UNIFIED PLATFORM EMPOWERING

KARPAGAM COLLEGE OF
177 CAREERS WITH REACT AND DJANGO 181
ENGINEERING
INTEGRATION
AI-BASED IT TRAINING SYSTEM:
FACILITATING EFFICIENT
COMMUNICATION AMONG TRAINING KARPAGAM COLLEGE OF
178 182
RESOURCES TO EMPOWER ENGINEERING
LEARNERS IN SKILL DEVELOPMENT

ONLINE XEROX AUTOMATION KARPAGAM COLLEGE OF

179 183
ENGINEERING
AVINASHILINGAM INSTITUTE
DRIVER DROWSINESS DETECTION FOR HOME SCIENCE AND
180 SYSTEM HIGHER EDUCATION FOR 184
WOMEN

DEEPFAKE DETECTION OF
M KUMARASAMY COLLEGE OF
SYNTHETIC MEDIA USING LSTM
181 ENGINEERING 185
ALGORITHM

PANIMALAR ENGINEERING
ANTISLEEP ALARM PROJECT –A
COLLEGE & KAMARAJ COLLEGE
182 PROBLEM SOLVING IDEA FOR 186
OF ENGINEERING AND
DRIVERS
TECHNOLOGY
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

REAL TIME BUS LOCATOR APPLICATION

Mr. B.Muthukrishna Vinayagam AP/CSE,
S.Subash, J.Subash Chandra Pandian
Department of Computer Science and Engineering,
Kamaraj College of Engineering andTechnology, Virudhunagar, Tamilnadu.

ABSTRACT:

In modern urban environments, public transportation systems play a vital role in

facilitating the movement of people efficiently and sustainably. However, challenges such as
unpredictable schedules and delays often deter individuals from utilizing these services to
their fullest potential. To address these issues and improve the overall user experience, the
Real-Time Bus Locator Application has been developed.This mobile solution harnesses the
power of Global Positioning System (GPS) technology to provide users with accurate and
real-time tracking of bus locations. Through a user-friendly map interface, individuals can
effortlessly monitor the exact whereabouts of buses within the transportation network. This
level of transparency empowers commuters to make informed decisions regarding their travel
plans, thereby reducing uncertainty and frustration associated with traditional bus
schedules.Key features of the Real-Time Bus Locator Application include the integration of
geofencing technology and push notifications. Geofencing enables the application to create
virtual boundaries around specific locations, such as bus stops or designated pickup points.
When a bus enters or exits these predefined areas, users receive timely alerts on their mobile
devices. This functionality ensures that individuals are promptly notified when a bus is
approaching their desired location, allowing them to prepare for boarding without
unnecessary waiting.Furthermore, the Real-Time Bus Locator Application prioritizes user
convenience by offering a seamless and intuitive interface. Navigation within the application
is straightforward, enabling users of all technical proficiencies to access the information they
need effortlessly. Whether planning daily commutes or spontaneous trips, individuals can
rely on the application to streamline their transportation experience and minimize the
uncertainties associated with public transit.

Keywords:GPS technology,User-friendly interface,Geofencing,Push notification.

1 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

AI-INFUSED WRIST COMPANION FOR MENTALLY

RETARDED HOSPITALIZED INDIVIDUALS
Suganya T, Swathi Sridhar, Abirami V, Jeeva Ganesh Kumar B, Harshathkumar B
Department of Artificial Intelligence And Data Science
KGISL Institute of Technology Coimbatore,India

ABSTRACT :
The healthcare sector places significant demands on medical professionals, often
resulting in considerable mental and emotional stress. This study addresses the widely
recognized issue of mental strain among healthcare workers and introduces a comprehensive
project aimed at mitigating these challenges through real-time monitoring. The proposed
project aims to improve the hospitalization experience for those with mental illnesses. This
cutting-edge wristband uses artificial intelligence to track and keep an eye on patients’ mental
health in real time. The device offers useful insights into the patient's mental health by
continuously gathering and evaluating data linked to emotional indications, behavioral
patterns, and vital signs. The wrist companion's monitoring capabilities are complemented by
its facilitation of communication and teamwork between the patient, healthcare professionals,
and caregivers. The device instantly notifies the authorized medical staff and designated
carers of any serious alterations or deteriorations in the patient's mental state through
intelligent notifications. The overall quality of life and recovery results for people with
mental disorders in hospitals are improved by the healthcare teams' ability to provide prompt
interventions and individualized care.

Keywords—Mental strain, hospitalization, companion, monitoring patients

2 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

ETHICAL CONSIDERATIONS IN SOCIAL NETWORK

ANALYSIS
Ms.V.Gayathri(Assistant Professor), Mr.S.Shiva Shankar, Mr.M.Logeshwaran,
Department of Information Technology, Kamaraj College of Engineering and Technology,
Virudhunagar.

ABSTRACT:
In the era of digital communication, Online Social Networks (OSNs) have become
integral platforms for social interaction. However, the proliferation of unwanted messages,
such as spam, hate speech, and irrelevant content, poses a significant challenge to users'
experience and platform integrity. This abstract proposes an intelligent filtering system
designed to mitigate the presence of undesirable messages on OSN user walls. One
fundamental issue in today On-line Social Networks (OSNs) is to give users the ability to
control the messages posted on their own private space to avoid that unwanted content is
displayed. Up to now OSNs provide little support to this requirement. To fill the gap, in this
paper, we propose a system allowing OSN users to have a direct control on the messages
posted on their walls. This is achieved through a flexible rule-based system, that allows users
to customize the filtering criteria to be applied to their walls, and a Machine Learning based
soft classifier automatically labeling messages in support of content-based filtering.
Additionally, the system integrates user behavior analysis to personalize filtering criteria
based on individual preferences and interactions. Through user feedback mechanisms and
historical data analysis, the system learns user preferences, allowing for fine-tuning of
filtering parameters to align with individual preferences and sensitivities.

Index Terms—Online Social Network (OSN), spam, hate speech, proliferation.

3 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

FOOD WASTE MANAGEMENT

Mrs.V. Gayathri(Assistant Professor), S. Veerasakthivel, T. Kowsalya
Department of Information Technology,
Kamaraj College of Engineering and Technology,Virudhunagar.

ABSTRACT :`
Minimizing food waste is crucial for ensuring sustainable resource utilization and
equitableaccess to nourishment, while also addressing logistical, safety, and cultural
obstacles. Thisabstract proposes an integrated approach to food waste management,
encompassing prevention, redistribution, recovery, recycling, and robust monitoring and
evaluation mechanisms. Prevention efforts prioritize initiatives aimed at reducing waste
generation through education, awareness campaigns, and behavioral interventions,
encouraging practices like meal planningand portion control. Establishing efficient surplus
food redistribution networks redirects ediblesurplus away from landfills to those in need,
fostering collaboration among stakeholders. Recovery technologies such as anaerobic
digestion and composting convert unavoidablefoodwaste into valuable resources like biogas
and organic fertilizer, necessitating investment andincentivization for scaling up. Promoting
recycling of food packaging materials andorganicwaste minimizes landfill contributions
through appropriate collection, separation, andprocessing systems. Robust monitoring and
evaluation frameworks track waste generation, diversion, and recovery metrics, utilizing data
analytics to optimize interventions and measureeffectiveness. This holistic approach
addresses the complex challenges of food waste whilefostering a resilient and sustainable
food system for all.

Keywords: surplus food redistribution, fostering collaboration, Minimizing food waste.

4 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

AI-ENABLED DISASTER MANAGEMENT SYSTEM :

ENABLING EFFECTIVE COMMUNICATION BETWEEN
RESOURCE AGENCIES TO SUPPORT PEOPLE DURING
CRISIS
DR.V.Rajeswari (Professor),Anand CT,Obuli Sridharan G,Gopala Krishnan S
Department of Computer Science andTechnology

Karpagam College of Engineering Coimbatore.

ABSTRACT:

In the dynamic landscape of disaster relief, effective coordination is important. This

paper introduces an innovative application designed to reshape how rescue agencies
collaborate and distribute resources during emergencies. At its core lies a centralized
database, the backbone of the system, storing crucial information from agency locations to
specialized skills. Utilizing intuitive interfaces and cuttingedge GPS technology, the app
provides a visually engaging map for easy agency identification and filtering based on
specific criteria. Yet, its impact extends beyond mere mapping. The application fosters real-
time communication and resource sharing among agencies, promoting efficiency and
synergy. Robust security protocols ensure data integrity and user privacy, building trust in the
platform's reliability.

Keywords— Geolocator , Triangulation Method, Location Detection , Navigation ,

Targeting

5 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

CYBER SECURITY
SHARWESH. S
Kongu Polytechnic College, Perundurai

ABSTRACT:

In today's hyperconnected digital landscape, cybersecurity stands as a crucial shield

against a myriad of evolving threats. This abstract encapsulates the essence of a presentation
by Sharwesh.S from Kongu Polytechnic College on Ensuring Cybersecurity in a Connected
World.The presentation dives into the multifaceted realm of cybersecurity, beginning with a
comprehensive definition highlighting its role in safeguarding digital systems, networks, and
data. It emphasizes the critical pillars of cybersecurity - confidentiality, integrity, and
availability - as the cornerstones for protecting digital information.Amidst the current cyber
threat landscape characterized by ransomware proliferation and supply chain vulnerabilities,
the presentation underscores the importance of robust cybersecurity measures. It elucidates
the dire consequences of cybersecurity breaches, ranging from identity theft risks to
reputational damage and legal implications.To mitigate these risks, the presentation advocates
for a holistic approach to cybersecurity encompassing multi-factor authentication, regular
software updates, employee training, secure backups, and endpoint protection. It delineates
the advantages and disadvantages of various cybersecurity strategies, ranging from strong
passwords and employee education to data encryption and network
segmentation.Furthermore, the presentation delves into emerging technologies in
cybersecurity such as Artificial Intelligence (AI), Blockchain, and Behavioral Biometrics,
while also addressing challenges like sophisticated cyber threats, skill shortages, and
compliance pressures.In conclusion, the abstract underscores the paramount importance of
cybersecurity in fostering trust, privacy, and resilience in the digital realm. It calls for
continuous innovation, collaboration, and proactive strategies to navigate the evolving threat
landscape and ensure a secure digital future.

6 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

DESIGN OF AN ASYMMETRIC CPW FED WIDEBAND

MIMO ANTENNA FOR 5G APPLICATIONS

KUMARESAN P 1, Dr.M.PARANTHAMAN 2
1) PG Student M.E- Applied Electronics ,
Department of Electronics & Communication Engineering,
Kongunadu College of Engineering and Technology,Tholurpatti-621 215, Trichy,.
2)Assistant Professor, Department of Electronics & Communication Engineering,
Kongunadu College ofEngineering and Technology, Tholurpatti-621 215, Trichy.

ABSTRACT:

This paper presents the design and optimization of an asymmetric coplanar

waveguide (CPW) feed for a wideband antenna system for 5G applications. The proposed
design aims to achieve enhanced bandwidth and impedance matching while catering to the
stringent requirements of 5G communication systems. Through a comprehensive simulation-
based approach, the performance of the asymmetric CPW feed is evaluated, considering key
metrics such as return loss, bandwidth, radiation characteristics, and performance parameters.
The designed antenna system demonstrates promising results, showcasing its potential for
fulfilling the demands of high-speed and reliable wireless communication in the 5G
application.

Keywords: Asymmetric CPW Feed, Wideband Antenna, 5g Applications, Micro Strip Patch
Antenna Impedance Matching, Bandwidth Optimization.

7 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

AN ECONOMICAL OFF-SEASON FARM EQUIPMENT

RENTAL PLATFORM USING REACT AND DJANGO
FRAMEWORK

Harishmaa C H, Monisha Poorani R ,Sathya Sri S, Sharmiga G

Computer Science and Technology,Karpagam College of Engineering ,Coimbatore.

ABSTRACT:

This paper introduces an innovative platform designed to optimize the utilization of

farm equipment during off-season periods, thereby addressing the economic inefficiency of
underused agricultural machinery. Developed using the React framework for the frontend and
Django for the backend, the platform facilitates a user-friendly interface and robust backend
services, creating an efficient marketplace for equipment rental. By enabling equipment
owners to list their idle machinery and farmers to rent them on an as-needed basis, the
platform aims to reduce the financial burden on farmers needing access to specific equipment
and provide an additional revenue stream for equipment owners. The system architecture
combines React's responsive user interface capabilities with Django's secure and scalable
backend, ensuring a seamless and secure user experience. Preliminary testing indicates a
positive response from the agricultural community, suggesting potential for widespread
adoption and significant impact on the agricultural economy. This paper details the
development process, system architecture, and initial feedback, alongside an economic
impact analysis, highlighting the platform's potential to revolutionize off-season equipment
utilization in farming communities.

8 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

MULTILINGUAL LEGAL AI ASSISTANT

Sowmya H ,Abinaya N, Mithun S, Sreedhar C, Sreejith R

Department of Artificial Intelligence And Data Science

KGISL Institute of Technology Coimbatore,India

ABSTRACT:
Legal complexities and language barriers often impede access to justice, particularly
for vulnerable populations of India. Obtaining basic legal knowledge and navigating the legal
system can be daunting, leading to disadvantage and exploitation without the assistance of a
Legal practitioner. This Can be critical for many people who are denied basic rights because
of Little to no awareness. Through the Advancement of Large Language model(LLM) and
NLP (Natural Language Processing) and harnessing these with fine Tuning the prompts and
description based on Indian Constitution law or IPC and Training with deep learning models
to understand the different languages in India and help through the necessary awareness. This
application provides rapid legal research in various language settings, accurate case
prediction, and expedited document analysis through rigorous training and optimization.
Additionally, our solution fosters seamless communication and collaboration between legal
practitioners and clients, facilitating informed decision-making and improving accessibility to
legalservicestoeveryone.

Keywords: Natural Language Processing(NLP), Legal , LLM , Fine-tuning.

9 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

HEALTH MONITORING SYSTEM FOR TYPE-2 DIABETES

Dr.A.Murugan,Karthick Raja K,Madheshwaran S,Karthikeyan S,Kavin V

Department of Electronics and Communication Engineering,
M Kumaraswamy College of Engineering,Karur.

ABSTRACT:
This paper presents a Health Monitoring System specifically designed for individuals
with Type-2 Diabetes. The system integrates various technologies to provide real-time
monitoring, analysis, and management of key health metrics, such as blood glucose levels,
physical activity, diet, and medication adherence. Through the use of wearable sensors,
mobile applications, and cloud-based platforms, the system enables continuous monitoring
and personalized feedback to help patients better manage their condition. Additionally, the
system incorporates machine learning algorithms to analyze data trends and provide
predictive insights, allowing for proactive interventions and improved long-term health
outcomes. The Health Monitoring System for Type-2 Diabetes aims to empower patients,
caregivers, and healthcare professionals with actionable information and support, ultimately
leading to better disease management and quality of life.

10 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

AGRI INTELLECT – EMPOWERING FARMERS WITH

SMART AGRICULTURAL SOLUTIONS
Mrs. J. Sathiya Jothi, K.S.Mahalakshmi, P.Hemapriyah, P.Kamali
Department of Information Technology
Anjalai Ammal Mahalingam Engineering College, Tirurvarur.
ABSTRACT :
Artificial intelligence has been categorized as a subfield of computer science where in
machines perform smart learning tasks with the help of data and statistical methods.
Agriculture is one of the oldest methods done by humans. It provides many crucial things like
raw materials, food, and employment. Due to the increasing population, it is the need of the
hour that the agriculture sector should increase production of resources to match the current
demand of the population. The user interface of Agri Intellect is designed for accessibility
and ease of use, catering to farmers across diverse scales and technological backgrounds.
Through intuitive dashboards and mobile applications, farmers can access critical information
and receive timely alerts, and make informed decisions to optimize their agricultural
operations. In summary, Agri Intellect represents a transformative approach to modern
agriculture. It offers farmers for practical tools and insights to address the complexities of
contemporary farming. Without relying on IoT technology, Agri Intellect demonstrates the
potential of AI-driven solutions to revolutionize agricultural practices. It ultimately
contributes to a more prosperous and sustainable agricultural sector. At present, methods used
by farmers are very traditional and insufficient to increased the demand. To match future
demand, our new innovative agriculture methos need to be adopted. Artificial intelligence in
Empowering Farmers with Smart Agricultural Solutions can enhance the prediction of yield.
The aim of proposed method is to predict the optimized crop yield by rainfall, quantity of
nitrogen, phosphate, potassium consumption in lands for both small scale and largescale
farmers for future opportunities for agriculture.

Keywords— Smart Agriculture, Crop yield prediction, Catboost algorithm, Prediction

11 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

DESIGN AND IMPLEMENTATION OF SECURITY

BASED PARKING SYSTEM FOR SMART CITIES
Mrs.C.Thilagavathi(Professor),Praveenkumar.U,Prithivraj.A,
Sabthagiri.V.M, Sudarsanbalaji.S
Department of Information Technology,
M.Kumarasamy College of Engineering,Karur.

ABSTRACT:

The vehicles have become the main means of transport today, and especially in the
metropolitan area where the requirements for public parking spaces are getting more and
more frequent due to growing popularity of the vehicles in the city. Thus, the parking
management of vehicles will become even more important. However, there are often some
disadvantages in the traditional parking area, such as time-consuming management, lack of
parking security, the inconvenience for finding a parking space and so on. In this paper, a
design of an efficient smart parking management system is proposed and implemented by
mainly using the image processing techniques, Arduino board hardware and electrical
apparatuses. The proposed system provides several appealing features, such as automatic
license plate recognition, the best recommended parking space reminder, antitheft parking
security, and real-time display of parking space status to further improve the above-
mentioned issues in the traditional parking area.

Keywords: Parking Space Optimisation, Security Enhancement, Smart Cities, Reservation

System

12 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

CAR CLUB RENTALS

Mrs.R.Mekala(Professor),Sivapandi.S,Sivabharath.B,Sriramana.S
Department of Information Technology,
M.Kumarasamy College of Engineering,Karur.

ABSTRACT:

The Car Rental System is being developed for customers so that they can book their
vehicles from any part of the world. This application takes information from the customers through
filling their details. A customer being registered in the website has the facility to book a vehicle
which he requires. The proposed system is completely integrated online systems. It automates
manual procedure in an effective and efficient way. This automated system facilitates customer and
provides to fill up the details according to their requirements. It includes type of vehicle they are
trying to hire and location. The purpose of this system is to develop a web site for the people who
can book their vehicles along with requirements from any part of the world. The first phase involves
the grouping of car rental locations into pools, allowing car rental locations within a pool to share a
fleet of vehicles. In the second phase, the types and quantities of vehicles to be acquired and
returned to the car manufacturer and the geographical redistribution of vehicles among pools over
the long-term planning horizon are defined for each pool. The final phase involves the daily
operations in which the deployment of the fleet within each pool among its locations is define.

Keywords:Booking Space Optimisation, Security Enhancement, Smart Cities, Reservation

System

13 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

INTELLECTUAL AGRIBOT FOR BANANA FARMING

Dr .K . Selva Sheela, Manojkumar D, Bala B , Trishala AG
Department of Artificial Intelligence And Data Science
KGISL Institute of Technology Coimbatore.
ABSTRACT:
The intellectual agribot for banana farming aims to develop a comprehensive chat
assistant for farmers who engage in cultivating banana plants as part of their regular or day-
to-day work. This chat assistant is designed to assist farmers primarily during specific
seasons, especially when facing challenges such as diseases or insects that affect plant
growth, particularly during adverse climatic conditions. The main objective of this project is
to create an agricultural chatbot that provides farmers with suggestions and guidance for
optimizing banana production, utilizing the capabilities of LLM (Large Language Model). In
the future, we plan to extend its functionality to include assistance for cultivating other herbal
plants and vegetables.

14 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

DETECTION OF TIME STAMPED SUSPICIOUS ACTIVITY

IN A SURVEILLANCE CAMERA FOOTAGE
Vijayalakshmi M, Aayisha J, Deepthi J, Sneha K
Department of Information Technology,
M.A.M College of Engineering and Technology (MAMCET), Trichy.

ABSTRACT:
The rise of surveillance systems has led to an increased demand for effective methods
of detecting suspicious activities in CCTV footage. In response, this study proposes a novel
approach utilizing Convolutional Neural Networks (CNNs) for suspicious activity detection.
CNNs have demonstrated remarkable success in various image-related tasks, making them a
suitable candidate for analyzing video data. By training the CNN on annotated datasets
containing instances of both normal and suspicious activities, the network can learn
discriminative features that distinguish between benign and potentially harmful behaviors.
The proposed system employs frame-byframe analysis, extracting spatial and temporal
features from video sequences to capture nuanced patterns indicative of suspicious behavior.
Additionally, the network undergoes optimization to enhance its ability to generalize across
diverse surveillance environments and adapt to varying lighting conditions and camera
angles. Evaluation results on benchmark datasets showcase the effectiveness of the CNN-
based approach in accurately identifying suspicious activities while minimizing false alarms.
This research contributes to advancing the capabilities of surveillance systems in proactively
detecting and responding to security threats, thereby enhancing public safety and security in
various settings.

Keywords— surveillance , Convolutional Neural Network , Support Vector Machine

15 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

ADVANCED REAL-TIME OBJECT DETECTION IN SMART

VIDEO USING EDGE GROUP ALGORITHMS
Ms.Suganya P(Assistant Professor ),Bhavani J,Jathika R,Sushma S
Department of Information Technology,
M.A.M College of Engineering and Technology (MAMCET), Trichy.

ABSTRACT:
Video surveillance systems are essential for maintaining security in a variety of settings,
including ATM locations. A key element of these systems is real-time object detection, which
permits prompt reactions to possible security concerns. This study suggests a brand-new
method for real-time object recognition designed especially for intelligent video monitoring
in ATM settings. Our approach reduces latency and bandwidth requirements by processing
video streams locally by utilizing edge computing capabilities. Our methodology is based on
the use of edge group algorithms, which allow for distributed processing among several edge
devices near the ATM. While preserving realtime performance, this distributed architecture
improves the system's scalability and resilience. The suggested method uses convolutional
neural networks (CNNs) that have been trained on massive datasets as part of a deep learning
technique for object detection. These models may be integrated into our edge group
framework to provide high detection accuracy for a variety of objects of interest, such as
people, cars, and suspicious objects. By employing anonymization techniques at the edge
nodes, you can further solve the privacy preservation difficulties and ensure that privacy
standards are followed while retaining the effectiveness of surveillance. The suggested
approach is evaluated by real-world video footage acquired from ATM locations throughout
the tests. The outcomes show how successful and efficient our method is, with low
processing delays and excellent detection accuracy.

Keywords: Real-Time, Object Detection, Smart Video, Computer Vision, Deep Learning,
User security

16 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

MULTI FACTOR AUTHENTICATION FOR WEB

APPLICATION USING MACADDRESS
E.Vakaimalar (Assistant Professor), S.Charles , J.BALAARJUN
Department of Information Technology
Kamaraj College of Engineering andTechnology
ABSTRACT:
Now a days every web application has a security mechanism that allows users
toaccesstheservices the web application has to offer. The authentication mechanism plays a
significant roleinprotecting user information. Many data leaks occur because users often
utilize easily guessableoruncomplicated passwords to safeguard their critical information.
Passwords are the first lineofdefense against unauthorized access and since they are at times
easily guessable, they are usedalongwith other authentication factors. They may be OTP,
Biometrics etc. forming a twofactorauthentication or multi factor authentication. Even though
they provide significant protection, theycan still be cracked. The proposed approach
introduces an authentication scheme that utilizesuser’sMAC address to provide security.
Along with this OTP and security questions act as additional lineofdefense when a user logs
in to the web app. Since MAC address is an important piece of informationSHA-256
algorithm is used to encrypt the MAC for secure storage. Our work helps toachieveasecure
way of authentication for users to access web applications that is efficient, andit
iscosteffective since it employs the usage of software token as a means of
authenticatingusers, thusproviding them with a seamless authentication experience.

Index Terms - MAC-Multi Factor Authentication, OTP – One Time Password , SHA-
SecureHashAlgorithm.

17 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

PSYCHE PLAYER
Emotion Based Music Player
Dr.E.Vakaimalar(Head of the Department), S.Ferdin Rodriguez, D.Sudhakar
Department of Information Technology
Kamaraj College of Engineering and Technology, Virudhunagar.

ABSTRACT:
In recent years, the convergence of facial emotion recognition technology with
music playback has emerged as a transformative frontier in human-computer interaction. This
report explores this intersection, delving into technological advancements, potential
applications, and societal implications. It discusses the evolution of facial emotion
recognition algorithms, personalized music experiences, and applications across
entertainment, healthcare, education, and human-robot interaction. While highlighting the
promise of these technologies in enriching lives and fostering deeper connections, the report
also addresses ethical concerns and the need for regulatory oversight

Keywords: Emotion detection, Song recommendation, Personalization, Neutral emotion

analysis, Emotional ambiguity

18 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

NON INVASIVEFETALMONITORING WITH COMBINED

PCG-PULSEMONITORING SYSTEM
Dr.P.Sutha,Abirami.S,Devisha.S , Kodeeswari.M
Department of Biomedical Engineering ,
PSNA College Of Engineering And Technology,Dindigul.

ABSTRACT:
In order to get the most recent information on the overall health of the foetus, it is
necessary to monitor the changes that occur in the physiological parameters of the organ.
The conventional method of fetal monitoring is mainly dependent on ultrasonic
methods; nevertheless, it is not suggested for the foetal health to be eśposed to ultrasound
radiation for an eśtended period of time or on a continuous basis. We intended to build a
digital fetal phonocardiogram in addition to a pulse monitoring apparatus in order to
circumvent this obstacle. Within this configuration, there are two sensors that are
incorporated: a pieioelectric sensor that is utiliied for the purpose of detecting the
heart sound, and a pulse sensor that is used for the purpose of aiding the recording of the
foetal pulse. The analog-to-digital converter (ADC) of the Arduino is then used to
process the signal that was detected. After that, the data that has been processed may be
sent to a terminal window so that the output can be shown. The information that is
collected from the sensors on the condition of the fetus is uploaded on a regular basis to the
Thingspeak server. This information is then accessible to the attending physician or
caregiver. Accessing the web server from a distant location allows the doctor to
remotely monitor the state of the foetus using this technology

19 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

COMPARATIVE ANALYSIS OF 3D ROOT CANAL MODELS

MRS.S.VIDHYA ,R.K.THARANI,K.SUBATHRA, N NIVETHA.
Department of Biomedical Engineering ,
PSNA College of Engineering And Technology, Dindigul.
ABSTRACT:
Root canal treatment is a common dental procedure that requires precision and
accuracy to ensure successful outcomes. Various factors, including the demands of dental
practitioners, improvements in dentistry education, and the availability of technology, can
influence the need for 3D root canal models. With the dentistry sector, it's a niche market
with potential value. The advent of 3D printing technology has revolutionized various fields,
including dentistry by enabling the production of customized surgical guides at lower costs.
This work is focused on a comparative analysis of 3D root canal models fabricated using
clear resin and polylactic acid (PLA) materials. Root canal therapy demands precise
understanding of root canal morphology, making 3D models crucial for education, research,
and clinical practice. Employing digital laser processing (DLP) for clear resin models and
Fused Filament Fabrication (FFF) for PLA models, the efficacy of the models is evaluated.
Both materials serve educational purposes well, with clear resin ideal for intricate
demonstrations and PLA for economical mass production. But the clear resin model has
superior surface finish and excels in accuracy than PLA models. From the proposed work it
is concluded that the clear resin model aids in treatment planning and patient communication,
while PLA remains valuable for preoperative assessment. This analysis guides dental
professionals in material selection, balancing precision with practicality to enhance
endodontic practice.

Index terms: Digital Light Processing (DLP) technology, Fused Filament Fabrication (FFF)
technology, Root canal Therapy (RCT), anatomical model, 3D printing.

20 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

SUSTAINABLE TOURISM DEVELOPMENTSYSTEM

Mrs.M Mythili,(Professor), Santhosh.C,Rishwanth.V.S,Ruban.A, Vishal.K.S.
Department of Information Technology
M.Kumarasamy College of Engineering,Karur.

ABSTRACT:
Sustainable tourism development is a comprehensive and dynamic system
designed to balance economic growth, environmental preservation, and socio-cultural
integrity. This approach emphasizes responsible and ethical practices that minimize
the negative impact of tourism on the environment and local communities while
maximizing positive contributions. Key components of this system include promoting
eco-friendly initiatives, conserving natural resources, supporting local economies, and
fostering cultural appreciation. By integrating sustainable practices into the tourism
industry, destinations can create a harmonious relationship between tourism and the
environment. This involves implementing strategies such as eco-certifications for
accommodations, promoting lowimpact transportation, and engaging in community-
based tourism initiatives. Education and awareness play crucial roles in this system,
encouraging both tourists and industry stakeholders to make informed choices that
prioritize long-term environmental and social well-being. Collaboration among
government bodies, businesses, and local communities is essential to implement and
enforce policies that support sustainable tourism development, ensuring a delicate
balance between economic prosperity and environmental stewardship.

Keywords:Local economies,Responsible practices,Sustainable tourism,Restoration.

21 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

ONLINE RESIDENCE RENTAL MANAGEMENT ZONE

Dr.R.Bharathi(Professor,), Santhosh.K, Nithish Kumar.S, Vijayakumaran.P.B.
Department of Information Technology,M.Kumarasamy College of Engineering, Karur.

ABSTRACT:
This paper presents a project called “Quick Homes” which is based on a house rental
management system for an online website. The project aims to provide a user-friendly
platform for property owners and tenants to interact with each other. The system allows
property owners to list their properties and manage bookings, while tenants can search for
properties, book them, and make payments online. The project is developed using PHP and
MySQL, and it includes features such as owner and tenant login, house details, booking
details, and member details. The project is designed to be simple, efficient, and secure, and it
can be used by property owners and tenants of all levels of technical expertise. In conclusion,
the Quick Homes project is a valuable contribution to the field of house rental management
systems. It provides a user-friendly platform for property owners and tenants to interact with
each other, and it includes features that make it easy to manage properties, bookings, and
payments. The project is developed using PHP and MySQL, and it is designed to be simple,
efficient, and secure. It is recommended for property owners and tenants who are looking for
a reliable and easy-to-use house rental management system.The core features of the "Online
Residence Rental Management Zone" encompass user-friendly interfaces for property listing
and search, robust communication channels for inquiries and negotiations, secure payment
gateways for rental transactions, and integrated tools for lease management and
documentation. Additionally, the platform incorporates advanced functionalities such as
automated reminders for rental dues, maintenance requests, and inspection schedules, thereby
enhancing operational efficiency and tenant satisfaction.

Furthermore, the "Online Residence Rental Management Zone" prioritizes data

security and privacy, implementing stringent measures to safeguard sensitive information and
ensure compliance with regulatory standards. Through the utilization of data analytics and
reporting tools, landlords and property managers can gain valuable insights into market
trends, tenant preferences, and property performance, enabling informed decision-making and
strategic planning.

22 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

Overall, the "Online Residence Rental Management Zone" represents a paradigm shift
in the realm of property management, offering a centralized solution that fosters
transparency, convenience, and productivity for all stakeholders involved. As the demand for
rental properties continues to surge, this digital platform stands poised to revolutionize the
way residential real estate is leased, managed, and optimized in the modern era.

Keywords:Online Residental Rental management,Digital Platform,Property

Listing,Communication Channels,Secure Payment Gateways.

23 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

INTEGRATING LANGUAGE TRANSLATOR TO ENHANCE

SECURITY AND ACCESS CONTROL USING API
Aishwarya K,Raksana P, Sneha K

Computer Science and Technology, Karpagam College of Engineering ,Coimbatore

ABSTRACT:
This app makes communication easier between people who speak different languages.
It allows users to input text and choose their desired languages for translation. With intuitive
text input fields and language selection dropdowns, the process is simple and seamless. The
app fetches accurate translations using the MyMemory Translation API, ensuring reliability.
Additionally, it integrates the Web Speech API for text-to-speech functionality, improving
accessibility for users with visual impairments or those who prefer auditory learning. Overall,
the app enhances real-time communication by providing instant translations. It bridges
language barriers and facilitates understanding among diverse language speakers. The user
interface is user-friendly, making it easy for anyone to use. Users can quickly type their
message and select the target language, and the app takes care of the translation. By
leveraging advanced technology, the app delivers precise translations, enhancing the user
experience. Moreover, the inclusion of text-to-speech capability adds another layer of
accessibility, catering to a wider audience. Whether for personal or professional use, this app
is a valuable tool for overcoming language barriers. It promotes inclusivity and fosters
communication across linguistic divides. With its efficient translation engine and user-
friendly interface, the app is a reliable companion for multilingual communication. Whether
for travelers, students, or professionals, this app simplifies language exchange and fosters
global connections. Additionally, its real-time translation feature ensures smooth and efficient
communication in various contexts. The app's reliance on the MyMemory Translation API
guarantees accurate and consistent translations. Furthermore, the integration of the Web
Speech API enhances accessibility by offering text-tospeech functionality. Overall, this app is
a powerful tool for breaking down language barriers and promoting cross-cultural
communication. Whether in personal interactions or professional settings, this app facilitates
seamless communication across languages. Through its innovative features and userfriendly
interface, the app empowers users to connect and collaborate regardless of language
differences.

24 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

EASY BUY
R.Haritha , B.Satya Meghana ,Dr.M.Saraswathi(Assistant professor)
Department of Computer Science and Engineering
SCSVMV University
Kanchipuram.

ABSTRACT:
An ecommerce website is a digital storefront on the internet. It is the virtual space
where you showcase products, and online customers make selections.It facilitates the
transaction between a buyer and seller. Ecommerce websites play a pivotal role in
fostering entrepreneurship and expanding market reach. Ecommerce websites
epitomize the convergence of technology , commerce and consumerism.

KEYWORDS:
Digital store front , Drop shipping ,E-business , marketing , Online security,
shopping, Shoppingcart.

25 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

A SMART SYSTEM FOR ACCIDENT AVOIDANCE USING

RAPID EYE MOTION DETECTION
P.Madhavan , K.R.Arun pragash , Dr.E.Emerson Nithiyaraj(AP)
Department of Electronics & Communication Engineering
Ramco Institute of Technology
Rajapalayam.

ABSTRACT:
This project offers a cutting-edge method of using fast eye motion detection
technologies to improve road safety. By continuously monitoring drivers' eye movements in
real-time and promptly issuing warnings or intervening when signs of tiredness or distraction
are observed, this technology seeks to reduce the number of accidents. The technology can
identify possible driver impairment by detecting quick eye movements, alerting the driver or
starting remedial action on its own to avoid collisions. This intelligent system provides a
preventative measure to lessen the dangers connected to driver weariness and inattention on
the road by integrating cutting-edge machine learning algorithms and sensor technology,
ultimately leading to safer driving environments.

26 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

DETECTION OF FETAL ANENCEPHALY IN BRAIN BASED

ON ENSEMBLE LEARNING
Durga S , Ashbinal Derija S
Department of Computer Science And Engineering
SCAD College of Engineering and Technology
ABSTRACT:
Identifying fetal mind irregularities from attractive reverberation imaging (X-ray) is
significant, as roughly 3 of every 1000 ladies are pregnant with a fetal of unusual mind. Early
discovery of fetal cerebrum irregularities utilizing machine learning methods can work on the
nature of analysis and treatment arranging. The writing has shown that a large portion of the
work made to group cerebrum anomalies in early age is for preterm newborn children and
youngsters not fetal. A pipeline process is proposed for fetal cerebrum irregularity location
which utilizations AI procedure. The fundamental commitment is the identification of fetal
mind abnormality (Anencephaly) in beginning phase, before the Fetal is conceived. The
proposed calculation is equipped for distinguishing irregularities from X-ray pictures with an
extensive variety of fetal gestational age (GA) (from 16 to 39 weeks) utilizing an adaptable
and basic technique with low computational expense. The novel proposed strategy comprises
of four stages; division, enhancement, feature extraction and classification. The Straight
separate analysis(LD), support vector machine (SVM), K-closest neighbour (KNN), and
Gathering Subspace Separates classifier are utilized to get better exactness.

KEYWORDS: Fetal Anencephaly , Brain Abnormalities , Ultrasound Diagnosis ,

Neurological Disorders , Healthcare Technology , Medical Imaging.

27 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

A NOVEL PARADIGM OF INDOOR NAVIGATION SYSTEM

USING LI-FI TECHNOLOGY
Aathiba D, Santhosh C
Second Year EEE, Second Year ECE
Knowledge Institute of Technology, Salem.

ABSTRACT :

According to a 2022 survey by the World Health Organization (WHO), nearly 2.2 billion
people across the globe are living with visual impairments, and this number is poised to
grow, especially as the global population ages. When these individuals visit unfamiliar indoor
locations, they face many difficulties. While navigation indoors is convenient for sighted
individuals, it is essential for the visually impaired to have aiding tools for their daily
activities. Electronic Travel Aids (ETAs) were developed to detect obstacles and guide users
through acoustic signals for simple navigation. To address these challenges, an indoor
navigation system utilizing Li-Fi technology was developed. This technology uses LED light
bulbs for data transmission and operates on a free, unlicensed spectrum that is unaffected by
RF noise. Li-Fi provides high-speed, low-cost wireless communication, is highly secure, and
cannot be easily intercepted. Additionally, it offers a larger bandwidth compared to Wi-Fi. To
enhance user convenience, the system is integrated with an ultrasonic sensor for obstacle
detection and an accelerometer to monitor the user's angular velocity. This integration makes
the system well-suited for guiding visually impaired individuals in unfamiliar indoor
environments.

Keywords-- Visually impaired people, Indoor navigation system, Light-Fidelity

technology, Light Emitting Diode, Wireless-Fidelity, ultrasonic sensor, accelerometer.

28 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

AUTOMATIC ROOM LIGHTS CONTROL BY USING

ARDUINO DESIGN
Dr.L.Lakshmanan,(Head department of CSE),Moirangthem Sanathoi Singh
Department of Computer Science And Engineering,
Sathyabama Institute of Science and Technology

ABSTRACT:
By recognising the presence of a human, an automatic light control system thatuses
an Arduino and a PIR (passive infrared) sensor may switch on and off the lighting ina
home. This system may be utilised in places like toilets, garages, classrooms, stairwells,and
other places where people aren't always present but still require light. Additionally,there is
no need to worry about power costs since, in the absence of a person, the lights goout and
one must pay the bills according to usage. This research proposes an automaticlight control
system that uses a PIR sensor and an Arduino to control thelights in theroom automatically.
This system's primary parts are an Arduino Uno, a PIR sensor, and arelay module. The PIR
sensor, which aids in the detection of human presence, is thecomponent that powers the
system the most out of all of them. The suggested techniquecansave
atleast50%ofelectricity,accordingtoexperimentaldata.

Keywords: Automatic room light, LDR sensor, PIR sensor, and Arduino

29 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

LEGALEASE CHATBOT – BRIDGING LEGAL

KNOWLEDGE GAPS FOR MARGINALIZED COMMUNITIES
Mrs.S.Sudha(Assistant Professor), Mr.K.Jothikumar, Mr.S.Aravind, Mr.R.Dhanush Raj,
Mr.K.Jaibharath
Department of Information Technology
Anjalai Ammal Mahalingam Engineering College, Tirurvarur.

ABSTRACT :

The Digital AI Assisted Chatbot for Legal Support in Marginalized Communities is a

pioneering solution designed to bridge the gap in access to justice for marginalized
individuals. By harnessing the power of artificial intelligence and chatbot technology, the
system aims to provide timely, accessible, and culturally sensitive legal support services.
Through features such as legal information dissemination, personalized advice, referrals to
legal aid organizations, and language accessibility support, the chatbot seeks to empower
marginalized communities to navigate the legal system effectively. This innovative solution
has the potential to promote social justice, equality, and empowerment for all members of
society. Overall, the implementation of such AIpowered chatbots represents a significant
advancement in democratizing access to legal information and supporting to develop a user-
friendly digital assistant providing legal information in multiple languages, aiding
accessibility and improving legal awareness among marginalized communities in India.

30 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

THREATS ANALYSIS AND DETECTION FOR DARKWEB

DATASET USING MACHINE LEARNING APPROCHES
Jebageetha.K,Arya.S , Kalaivani .S, Micheal Jerlin.V
Department of Computer Science and Engineering
JP College Of Engineering

ABSTRACT:
This study investigates the analysis and detection of threats within Dark Web datasets
using machine learning approaches. Specifically, the research focuses on identifying and
categorizing various types of attacks, including Denial of Service (DoS), Probe, and
Ransomware, prevalent in Dark Web forums and marketplaces. By employing machine
learning algorithms, the study aims to develop robust detection mechanisms capable of
effectively identifying and mitigating potential threats lurking within the hidden corners of
the internet.The Dark Web presents a clandestine environment where cybercriminals operate,
exchanging illicit goods and services, including tools for cyber attacks. Traditional security
measures often struggle to monitor and detect malicious activities within this hidden realm
due to its anonymity and encryption features. Consequently, there is a pressing need for
advanced threat detection techniques that can analyze Dark Web datasets and identify
emerging cyber threats, such as DoS, Probe, and Ransomware attacks.

The primary objective of this research is to develop a comprehensive threat analysis and
detection framework for Dark Web datasets, focusing on the identification of DoS, Probe,
and Ransomware attacks. Through the application of machine learning approaches, including
supervised and unsupervised learning algorithms, the study aims to classify and characterize
various attack vectors present within the Dark Web ecosystem. By achieving this objective,
the research seeks to enhance cybersecurity measures and preemptively counteract potential
threats originating from the Dark Web.

The implementation of machine learning approaches for threat analysis and detection within
Dark Web datasets yields promising results. Through the utilization of supervised learning
algorithms such as Random Forest and Support Vector Machines, along with unsupervised
learning techniques like clustering, the system effectively identifies and categorizes DoS,
Probe, and Ransomware attacks with high accuracy. By analyzing patterns and anomalies
within the data, the machine learning models successfully differentiate between normal and

31 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

malicious activities, enabling proactive threat mitigation strategies .In conclusion, the
utilization of machine learning approaches for threat analysis and detection within Dark Web
datasets offers a potent solution for enhancing cybersecurity measures. By accurately
identifying and categorizing various attack vectors, including DoS, Probe, and Ransomware,
the developed framework enables organizations to fortify their defenses against emerging
cyber threats originating from the Dark Web. Moving forward, continued research and
refinement of machine learning-based detection techniques will be crucial in staying ahead of
cyber adversaries and safeguarding critical digital assets and infrastructure.

32 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

TOXIC COMMENTS CLASSIFICATION USING MACHINE

LEARNING
Kandhasamy V1, Sriram K2
1)B.Tech IT, Paavai Engineering College(Autonomous), India
E-mail: [email protected]
2)B.Tech IT, Paavai Engineering College(Autonomous), India
E-mail: [email protected]
ABSTRACT:

The proliferation of online platforms has facilitated widespread communication but

has also led to the emergence of toxic comments, which can propagate hate speech,
harassment, and misinformation. In this abstract, the present a study on leveraging machine
learning techniques for the classification of toxic comments. We begin by assembling a
diverse dataset of comments annotated for toxicity, encompassing various forms of harmful
behavior. Utilizing this dataset, we explore a range of machine learning algorithms, including
traditional classifiers and deep learning models, to develop effective toxicity detection
systems. We investigate feature engineering techniques such as TF-IDF, word embeddings,
and contextual embeddings to capture semantic and syntactic nuances in toxic language.
Additionally, we address challenges such as dataset imbalance through strategies like
oversampling, under sampling, and costsensitive learning. Through rigorous experimentation
and evaluation, we assess the performance of our models in terms of accuracy, precision,
recall, and F1- score. Our results demonstrate the efficacy of machine learning approaches in
accurately identifying toxic comments across diverse online platforms. By automating the
identification of toxic content, our research contributes to fostering healthier online
communities, empowering platforms to proactively moderate harmful behavior and promote
positive discourse.

Keywords: Leveraging, TF-IDF, rigorous, syntactic nuances.

33 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

SPECTRAL SUBTRACTION ALGORITHM FOR AUDIO

NOISE REDUCTION: A COMPARISON WITH KRISP
A. Arul Xavier,S. Ram kumar, V. Matheesh kumar
Department of Computer Science and Engineering
JP College Of Engineering
ABSTRACT:

This paper investigates the effectiveness of the Spectral Subtraction Algorithm

(SSA) as an alternative to the widely used Krisp algorithm for noise suppression in audio
signals. A literature review will analyze the limitations of Krisp, while exploring the potential
of SSA for noise reduction. The paper details the implementation of SSA and describes the
experimental setup, including the selection of audio datasets, noise types, and evaluation
metrics. The performance of SSA will be compared with Krisp using both objective and
subjective evaluation metrics. The study aims to demonstrate that SSA achieves superior
noise reduction effectiveness compared to Krisp, while maintaining reasonable computational
complexity. This research contributes to the exploration of alternative noise suppression
techniques and provides valuable insights for further development in this domain. Greater
noise reduction By directly targeting specific noise frequencies, SSA might achieve a higher
degree of noise removal compared to Krisp's broader noise reduction techniques.
Computational efficiency Depending on the implementation, SSA might require less
processing power than Krisp, making it suitable for real-time applications on resource-
constrained devices.Improved robustness SSA might handle specific noise types, like white
noise, more effectively than Krisp, leading to cleaner audio output in such scenarios.
Comparison with Krisp: The core of the study involves a head-to-head comparison of SSA
with Krisp. The paper expects to demonstrate the superiority of SSA through Objective
Evaluation is Higher SNR improvement achieved by SSA across different noise types and
datasets. In Subjective Evaluation Listeners perceiving audio processed by SSA as cleaner
and easier to understand compared to Krisp. In Conclusion, This study investigates the
potential of Spectral Subtraction Algorithm (SSA) as a superior alternative to the Krisp
algorithm for audio noise reduction. By achieving greater noise reduction effectiveness,
improved computational efficiency, and potentially better robustness against specific noise
types, SSA could become a valuable tool in enhancing speech communication quality than
Krisp Algorithm.

34 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

FREEBEEZ: A FREELANCERS COMMUNITY WEBSITE

Dr.S.Geeitha 1 ,Paramesh.M 2 ,Saran.S 3 ,Sudarsan.D.S 4 ,Surya.S 5

1) Professor, Department of Information Technology,
M.Kumarasamy College of Engineering,Karur,TamilNadu,India-639113
2,3,4,5 UG Students, Department of Information Technology, M.Kumarasamy College of
Engineering,Karur,TamilNadu,India-639113 1 [email protected]

ABSTRACT:
The Freelancer Community Website, documented herein as Project FREEBEEZ, is a
forward-thinking online platform poised to redefine the landscape of freelance collaboration.
In response to the increasing prominence of freelancing in today's workforce, the platform
has been meticulously designed to serve as a dynamic hub where freelancers and clients
seamlessly connect. Emphasizing the project's scope and its potential transformative impact,
this section sets the stage for understanding the multifaceted nature of Project FREEBEEZ.

The frontend and backend technologies, database schema, and security measures are
illuminated, providing insights into the mechanisms that underpin the platform's
functionality. The discussion extends to user roles, permissions, and future scalability
strategies, highlighting the adaptability of Project FREELANCER to meet the evolving needs
of the freelance industry. The abstract serves as a comprehensive guide, encapsulating the
essence of Project FREEBEEZ objectives, features, technological foundations, and potential
impact on the freelance landscape. Within the multifaceted realm of Project FREEBEEZ, the
detailed user guide offers practical insights on navigating the platform's functionalities.
Users, comprising freelancers, clients, and administrators, are provided with comprehensive
instructions on registration, profile management, job postings, bidding on projects, utilizing
the messaging system. Each section is crafted with clarity, ensuring users can harness the
platform's capabilities effectively.

Keywords:Freelancers,Online platform,Connection,Freelancing

35 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

AUTOMATED SMOKE AND GAS LEAKAGE DETECTOR

WITH IOT MONITORING SYSTEM
Vallinayagam.N,Natarajan.S,Thirumalaikumar.K,Arumuganathan.K
Department of Computer Science and Engineering
JP College Of Engineering

ABSTRACT:
The majority of explosions are triggered by gas leakage that goes unnoticed in the
pre-detection stage. A gas leak resulted in a tragic accident because it has optimal properties
such as high heat value, less smoke, less soot, and less environmental impact, liquefied
petroleum gas (LPG) or propane. The objective of this project study is to design a tool using
technology and the internet of things (IoT) that functionally detects and prevents wireless
threats. A device made from microcontroller Arduino UNO that controls all activity of the
device alongside node WiFi module, nodemcu ESP8266 that connected and communicated
using the internet to send a notification warning to users. The device's efficiency is examined
in this study by running multiple trials to see how well the sensor responds to gas or smoke
leaks. The system displayed a success rate of 25 out of the total of 30 repeated measurements,
giving the device efficiency of 83 percent. This project demonstrates that it is able to detect
and monitor gas leaks in a reasonably enclosed space. The module may be used as a
household application to monitor and take appropriate control actions in various conditions.
Gas leaks cause serious accidents that result in property damage and human injuries. The
major causes of gas leaks are poor equipment maintenance and a lack of public awareness. As
a result, LPG leaks detector is critical for avoiding accidents and saving human lives. This
strategy provides real-time data to the user, allowing them to be informed at any moment.

Keywords: Automated Smoke, Gas Leakage Detector, IoT, Monitoring System

36 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

BLOODBANK MANAGEMENT SYSTEM

Mr. Kalaiarakan.k, AP/IT,1.Siva Prakash.R.S,2Shanmugapriyan.N,3Tamizhannal.M,
4
Sunthar shree.L
1
Professor, Department of Information Technology, M.Kumarasamy College of
Engineering,Karur,TamilNadu,India-639113
2345
UG Students, Department of Information Technology, M.Kumarasamy
College of Engineering,Karur,TamilNadu,India-639113

ABSTRACT:
A Blood Bank Management System (BBMS) is a platform that connects blood donors
and doctors for blood donation, ensuring that every patient can receive the required blood
within a timely manner. The BBMS aims to bridge the gap between donors and doctors,
providing a convenient platform for donors with busy schedules to donate blood. The system
includes features such as donor registration, storage of physical and medical details, and
eligibility checks for donors. The transfusion management module prioritizes patient safety,
implementing compatibility checks and electronic documentation for streamlined processes.
Additionally, the IBMS features analytics and reporting tools that provide valuable insights
into blood utilization patterns, aiding in data-driven decision-making. Forecasting tools are
integrated to predict future demand, optimizing inventory planning. The system places a
strong emphasis on regulatory compliance and security, adhering to standards for blood
banking and safeguarding sensitive patient and donor data.Regular updates ensure that the
BMS remains aligned with evolving healthcare standards, making it a dynamic and future-
ready solution. In summary, the IBMS offers a comprehensive and integrated approach to
blood supply chain management, promoting efficiency, transparency, and accountability in
healthcare institutions, ultimately improving patient care and resource utilization.

Keywords: Blood Donor,Getdetails,Interaction between donor and searcher

37 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

ACCURATE OVARIAN CYST CLASSIFICATION WITH

ALIGHTWEIGHT DEEP LEARNING MODEL FOR
ULTRASOUND IMAGES
S. Harsini, C. Sneka,
Information Technology,
Paavai Engineering College, Namakkal,Tamilnadu

ABSTRACT:

This project focuses on enhancing the accuracy of ovarian cyst classification through
the implementation of a lightweight deep learning model specifically designed for ultrasound
images. Ovarian cysts are a common medical condition, and accurate classification is crucial
for timely diagnosis and treatment. The model is fine-tuned on a dataset comprising a diverse
range of ultrasound images depicting ovarian cysts, encompassing various sizes, shapes, and
structural characteristics.The primary goal of this project is to create a highly accurate and
efficient deep learning model for classifying ovarian cysts in ultrasound images.This project
addresses a critical need in the medical field, where the early detection and classification of
ovarian cysts can significantly impact patient outcomes. The Image processing, ultrasound,
Deep Learning are the technologies used in this project. In Existing System Ultrasonography is
a common method for screening ovarian cycle. However, the demand for ultrasound has
exploded in recent years, doctors’ workloads have undoubtedly increased. So, in our proposed
system having the ultrasound image analysis of ovarian cysts using deep learning is aimed at
assisting doctors in rapid diagnosis and providing a good diagnostic decision for patients. The
Classification accuracy of the network is 95.93%, which has certain practicability in clinical
application.

Keywords: Deep learning, cyst classification, Ovarian, ultra sound.

38 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

IOT BASED SMART NEBULIZER

Mrs.U.Rujeyanthi,k.Saiharini,B. Sathya priya,A.Selvi
Department of Biomedical Engineering
P.S.R Engineering College,Sivakasi.

ABSTRACT:
The purpose of the present work is to develop an automated jet nebulizer with
dynamic flow regulating capabilities.With current technology, a significant portion of the
aerosol is released into the atmosphere through the vent, when the respiratory process is
exhaling. To be precise, half of the aerosol created volume is squandered.If this ineffective
delivery method is ignored, nebulization may not have the desired benefits.Adverse effects
such as bronchospasm and exposure to elevated medication concentrations could occur.The
"Compressed Air" and "Oxygen Therapy" modes make up the two modes of the suggested
nebulizer.The patient's oxygen saturation percentage, which is tracked by the SpO2 sensor,
With the use of an algorithm based on a temperature sensor, the patient's minute ventilation
will be used to determine how much compressed airflow to give. The controller for the
compressor circuits makes certain that the patient receives compressed air at the system-
determined flow rate. The nebulization procedure will stop at the end of the drug delivery if
the liquid level sensor notices that there is no medicine left in the nebulizer chamber. It was
successfully possible to dynamically regulate the motor speed in relation to the minute
ventilation. As anticipated, there was a turbulent flow inside the chamber and a laminar flow
from the compressor's output towards the nebulizer tube. There were no discernible circular
flow patterns or excessively turbulent flows. The outcome might undoubtedly result in
nebulizer enhancements over the current models.

39 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

LEVERAGING YOLO NAS AND CONTRASTIVE LEARNING

FOR NEXTGENERATION RETAIL ENVIRONMENT
ANALYSIS
Ms. P.Dhivya1 , Ms.M.Ramya2
1 Assistant Professor, Department of Artificial Intelligence and Data Science,
Karpagam College of Engineering, Coimbatore
[email protected]
2 Assistant Professor, Department of Information Technology,
Velalar College of Engineering and Technology, Erode
[email protected]
ABSTRACT :

In today's information era, online stores seem to know the customer better than retail
stores through advanced data analytics and tracking mechanisms. E-commerce have gained
great insights into customer details such as their last purchase, customer preferences,
shopping habits, and decision-making processes. To compete with the analytical potential of
online stores, retail stores can indeed leverage YOLO-NAS (You Only Look Once Neural
Architecture Search) object detection methods to better meet customer needs by enabling
computers to interpret and understand visual information, just as humans do. The main aim of
this paper is to detect and analyze customer behavior from video footage acquired by
surveillance cameras by combining YOLO-NAS object detection algorithm and contrastive
feature learning. YOLO-NAS, which is a new object detection model that is more efficient
and accurate than previous versions of YOLO. It works by using a neural architecture search
(NAS) technique to automatically design the network architecture. This means that YOLO-
NAS can find the best network architecture for a particular task without any human
intervention. By exploiting contrastive feature learning, the system can extract selected
features from customer interactions captured by the scale-invariant object detection process.
This allows for a more fine understanding of customer behavior beyond simple object
detection. By accurately detecting and tracking customers at different scales provide detailed
analysis of customer behavior such as store entry and navigation pattern, in-store dwell time,
purchase decision, influence of promotions and marketing. These details will help the retail
owners to make data driven decisions based on historical data and patterns such as optimize

40 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

store layouts, product placements, marketing strategies, estimating busy hours, optimize the
staff allocation, etc., leading to enhanced better customer experiences.

Keywords - Object detection, YOLO-NAS (You Only Look Once Neural Architecture
Search), Customer behavior, Contrastive feature learning, Data driven decision.

41 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

ALUMNI WEB PORTAL EMERGED WITH DEEP

LEARNING
Dr. Kalpana S1 , Jabin S2 , Rathna Praba N3
1Professor, 2,3UG Scholars
Department of Artificial Intelligence and Data Science, KGISL Institute of Technology,
Coimbatore.
ABSTRACT:
In recent years, alumni portals have emerged as essential tools for student
registration and fostering connections between colleges and alumni. However, many of these
platforms lack interactive elements such as chatbots, blog posts, and community events. To
address this gap, a novel approach harnesses AI technology to enhance the functionality of
these portals. By integrating features like chatbots, sentiment-enabled blog posts, job postings
with real-time notifications, and advanced search capabilities, the objective is to elevate the
user experience and increase engagement on these platforms. The solution adopts the Laravel
framework for web development and implements Artificial Neural Networks (ANNs) using
Keras, a user-friendly neural networks API based on TensorFlow, to enhance the chatbot
feature. Leveraging ANNs with distinct layers for processing input data and generating
output significantly enhances the chatbot's effectiveness in tasks such as classification and
pattern recognition. The combination of Laravel and ANNs through Keras ensures a
sophisticated and efficient web application, particularly in enriching user interaction through
chatbot functionality. By keeping users updated on real-time events and activities, the chatbot
promotes greater engagement. Furthermore, the approach has demonstrated improved project
efficiency, laying a robust foundation for alumni support and engagement.

Keywords: Alumni support, Artificial Neural Networks (ANNs), chatbots, connections,

Laravel framework

42 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

FRICTION STIR WELDING OFAA2219ALUMINIUM ALLOY

WITH THREE DIFFERENT PROFILE TOOLS AND THEIR
CHARCTERIZATION
Mr.P. Srinivasan ¹ , T. Ivon Derek ², S.J. Santhosh ³, G. Vijayendran ⁴,D. Gokul⁵
M. Harish ⁶, V. Jayasuriya ⁷.
¹,²,³,⁴,⁵,⁶,⁷, Department of Mechanical Engineering,
Periyar Maniammai Institute of Science and Technology,
Vallam, Thanjavur, Tamil Nadu, India- 613 403.
*Corresponding author, Email: [email protected]
ABSTRACT:
Friction Stir Welding has emerged as a transformative green technology, used in
aviation, automotive, railway and naval manufacturing industry to join high strength
aluminium alloy that eliminates the liquation and subsequent solidification cracking problems
associated with fusion welding process. AA2219, a high-strength aluminium-copper alloy
widely employed in aerospace and structural applications, known for its exceptional
cryogenic properties with combination of strength, toughness, and corrosion resistance. This
study focuses on the FSW of AA2219 aluminium alloy and the subsequent characterization.

This research aims to investigate the influence of three flat pin profile tools of square,
hexagonal, octagonal tools welding on AA2219 alloy with process parameters such as
welding traverse speed, tool tilt angle and tool rotational spindle speed. The after-weld
mechanical, microstructural and metallurgical properties of AA2219 aluminium alloy welds
were investigated. The research outcome shows, significant refinement of the grain was
identified in the WNZ due to intense plastic deformation, confirmed by EBSD analysis
showing finer grain sizes compared to the BM. Coarser grains were found in the HAZ due to
thermal cycling during FSW. The comparatively best quality weld results were obtained
when utilizing the square pin tool profile, as opposed to the octagonal and hexagonal profiles.

Key words: Friction Stir Welding, AA2219 aluminium alloy, square, hexagonal, octagonal pin
profile tools, mechanical, microstructural, EBSD characterization.

43 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

THE HYBRID MODEL FOR OPTIMIZED EXTREME

HEATWAVE PREDICTION USING ADAPTIVE NEURO-
FUZZY INFERENCE SYSTEMS
Kaviya P, Karthikeyan M, Srienath Vaisakh M
Department of Information Technology,
Kamaraj College of Engineering and Technology, Virdhunagar

ABSTRACT:
A natural disaster is an event that includes extreme climatic events like precipitation,
cyclones, floods, heatwaves, wildfires, volcanic eruptions and tsunamis which causes
significant damage, destruction, or loss of life. These disasters are caused by change in
temperature, humidity, air density, wind speed and direction, pressure. Recently, the various
extreme climatic observations generate a vast amount of data ‘big data’ from multiple sources
that can be used to forecast the extreme climatic events. But the conventional forecasting
system called Numerical Weather Prediction (NWP) lacks inference about the effective
attributes, many researchers are turning to deep learning algorithms to develop an accurate
and effective extreme climatic event forecasting system. Deep learning (DL) can offer a
flexible way of processing ample data and provides scalable solutions for data driven
applications. In this work, a Deep Learning based Weather Prediction (DLWP) approach is
used for forecasting the occurrence of heatwave. A deep ensemble model is proposed that
combines Artificial Neuro Fuzzy Interference System (ANFIS) to predict the occurrence of
heatwave. This model is trained using two different datasets include Indian Meteorological
Department (IMD) data set and European Centre for Medium-Range Weather Forecasts
(ECMWF) - ERA5 data set. As a result, this model able to predict daily maximum
temperatures over a range of steps from 1 day to 5 days with root mean squared error of
0.054.

Keywords: Natural Disasters, Heatwave Prediction, Numerical Weather Prediction, Deep

Learning based Weather Prediction, Artificial Neuro Fuzzy Interference System

44 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

DESIGN AND ANALYSIS OF A EQUAL SPLIT COMPACT

WILKINSON POWER DIVIDER FOR WIRELESS
COMMUNICATION SYSTEMS
V.Priyadharshini*, L.Tifinimaria, V.Uvaiz, G.Soundaraya
Department of Electronics and Communication Engineering,
Dr.Mahalingam College of Engineering and Technology, Pollachi, TamilNadu
*[email protected]
ABSTRACT:
The power divider is a prominent N-ports device commonly used in the microwave
and radio frequency applications. Generally, Wilkinson Power Divider (WPD) is the most
wellknown and respected of all power dividerswhich can be used extensively in balanced
power amplifiers, and radar systems.This divider uses a single frequency band and has two
lines with a length of L= λ/4 (quarter wave).These microwave devices are mainly designed to
establish a power distribution from an input signal to a predetermined number of output
signals. In this paper, WPD with two output ports has been designed and scattering
parameters were analyzed through ADS simulator. From the obtained results, it is found that
the divider operates between the S band frequency ranges (2.0 GHz to 2.5 GHz) which can be
suggested for the applications related to WLAN which are functioning at 2.4 GHz.

45 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

ADVANCED ADAPTIVE CRUISE CONTROL SYSTEM

Corresponding Author : P MEIYAZHAGAN B.E
Guide : Professor. Dr. ASHOK KUMAR M.E., Ph.D.
Department of Computer Science & Engineering,
Sathyabama Institute of Science and Technology,Chennai,India -600119

ABSTRACT:
This paper presents the development and implementation of an advanced adaptive
cruise control (ACC) system in an autonomous vehicle. The proposed system aims to achieve
efficient travel by managing speed, distance, start time, and end time based on user input and
real-time environmental conditions.This technology aims to increase the autonomy and
usefulness of cars through intelligent automation. The brains behind the seamless data
interchange between various components of the Internet of Things are the IoT
microcontrollers. Ultrasonic sensors supply critical data that is necessary for real-time
environmental awareness, enabling the system to modify its direction and speed in reaction to
its environment. The motor driver and DC motor that make up the propulsion system allow
for exact control over the vehicle's motions. The addition of a robot module improves the
system's flexibility by enabling it to navigate over complex terrain and adapt to changing
conditions. A unique battery management system extends the life of the vehicle by
guaranteeing optimal power consumption. In addition to promising a better economy, this
state-of-the-art cruise control system has an adaptive function that enables it to respond
appropriately to changes in the surrounding environment. The Adaptive Cruise Control
System is a significant development in intelligent transportation systems, paving the way for
safer, more efficient, and versatile automobiles in the era of smart mobility. It can traverse
challenging terrain, avoid obstructions, and adjust speed.

Keywords— Adaptive Cruise Control System, intelligent transportation systems,

automobiles.

46 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

FABRICATIONANDMECHANICAL

BEHAVIOROFAL20CO20FE20NI20MN20HIGHENTROPY ALLOY

THROUGH MECHANICAL ALLOYING

D.Jeyasimman1*,R.Udhayasankar2,M.TamilSelvan3,S.Logeshwaran4,M.Kanagarajan5
1, 2 ,3, 4& 5
Department of Mechanical Engineering,

Periyar Maniammai Institute of Science &Technology, Thanjavur.

*CorrespondingAuthor:Dr.D.Jeyasimman,[email protected];[email protected]

ABSTRACT:
This research work aims to synthesis and mechanical behavior of new advanced high
entropy alloys (HEAs) through mechanical alloying. The raw materials such as 40 µm
powders were milled up to 30 h by High Energy Ball Milling. Consolidation and sintering of
HEA powders done by two conventional sintering temperatures at 600° and 800° C. The
mechanical properties analysis of prepared high entropy alloy was done by using density,
hardness measurements and compression test. Density measurement was done by using
Archimedes principle.Vickers micro hadness testing was done with load 300g and dwell time
15s. Ultimate compressive strength of prepared high entropy alloy composite were
investigted. The proposed high entropy alloy Al20Co20Fe20Ni20Mn20 provides ultra high
power density, long life span and rapid energy transfer. Hence, this HEA has large energy
storage capacity so that used as thermo electric materials energy storage.

Keywords : High Entropy Alloy; Sintering; Density; Hardness; Compressive strength

47 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

SUSTAINING TRADITION: CHICKEN RANCH ORGANIC

E-COMMERCE EVOLUTION
T.Saravanan1, T.Thillai 2, S.Shaliha3, P.Sumithra4, M.Rajasri5
1
Asst. Prof., Department of IT, JP College of Engineering, Tenkasi, Tamil Nadu, India
2,3,4,5
UG Students, Department of CSE, JP College of Engineering, Tenkasi, Tamil Nadu.
Corresponding Author: [email protected]

ABSTRACT:

In the rapidly evolving landscape of e-commerce and sustainable agriculture, the

journey of Chicken Ranch presents a compelling case study. This conference paper delves
into how Chicken Ranch, a stalwart in organic poultry products, has successfully intertwined
tradition with modernity through its e-commerce evolution. By meticulously preserving its
traditional farming practices while embracing the digital realm, Chicken Ranch has carved a
niche for itself in the organic food market. This paper explores the strategic initiatives
undertaken by Chicken Ranch, the challenges encountered, and the transformative impact of
its organic e-commerce evolution on sustainable agriculture.

Keywords: Chicken Ranch, organic e-commerce, tradition, sustainable agriculture, digital

transformation

48 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

AGRI-KET(LIVE STOCK MANAGEMENT WEB PAGE)

RagulRaja C, Saran R, Sivaraj T

Department of Information Technology, M. Kumarasamy College of Engineering

ABSTRACT:

Agriculture remains the cornerstone of our nation's economy. However, contemporary

challenges have dissuaded many young individuals from engaging in agricultural pursuits due
to concerns about potential losses. In response, we present AGRI-KET, a comprehensive
livestock management platform. AGRI-KET comprises four primary modules: 1. Livestock
Grooming Module: This module meticulously catalogs data on livestock, encompassing
optimal maintenance practices, disease possibilities, and vaccination schedules from birth to
maturity. 2. Veterinary Outreach Module: Farmers can effortlessly connect with veterinary
professionals in their localities by simply specifying their district, facilitating timely and
informed livestock healthcare. 3. Direct Farmer-to-Consumer Sales: Empowering users to
purchase livestock directly from farmers, bypassing intermediaries. Each transaction is
accompanied by a health certificate, ensuring transparency and quality assurance. 4.
Livestock Marketplace: Utilizing both traditional and auctionbased sales mechanisms, this
module facilitates efficient livestock transactions, catering to diverse preferences and market
dynamics. AGRI-KET stands as a transformative solution, bridging the gap between modern
technology and age-old agricultural practices, revitalizing interest and confidence in
agriculture among the youth while enhancing efficiency and sustainability in the sector.

49 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

IOT BASED FARM PROTECTION SYSTEM

Mrs Saranya vaishalini V G,Gayathri P,Lavanya V

Department of Computer Science and Engineering,P.S.R Engineering College.

ABSTRACT:

This project aims to enhance farm protection by leveraging the Internet of Things
(IoT) and machine learning technologies. An IoT system will be implemented with sensors
strategically placed across the farm to monitor and collect data on animal movements.
Machine learning algorithms will then be applied to analyze this data, distinguishing between
normal and potentially harmful animal activities. The system will provide real-time alerts to
farmers, enabling prompt intervention to protect crops and livestock. Additionally, the
collected data will be utilized to continuously improve the machine learning models, ensuring
enhanced accuracy and effectiveness in farm protection over time. This innovative approach
integrates technology to create a proactive and efficient solution for mitigating potential risks
posed by wildlife to agricultural assets.

50 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

EXAMINATION HALL ALLOCATE

Srivarshini R,Sivapriya S , Sasikumar .S

Department of Electrical and Electronics Engineering,
DR Mahalingam college of engineering and technology

ABSTRACT:
EXAMINATION ROOM GUIDE USING RFID, presents a modernized method
of examination hall management. It is possible for a student to identify the
particular exam hall from any other hall, when they swipe RFID card in a card reader located
there. This helps them to identify the floor or get directions to their respective halls without
delays. The card reader is provided at the entrance of the building, if the students enters
wrongly a buzzer alarm sets off, otherwise the room number is displayed on the LCD,
connected to controller.

51 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

COVER TIMING CHANNELS USING IMAGE PROCESSING

AND MACHINE LEARNING
Mrs. Suguna Angamuthu(Assistant Professor), Arun V, Gowtham Kumar G, Runil V
Department of Information Technology, Nandha Engineering College,
Erode.

ABSTRACT :

The increasing sophistication and use of Covert Timing Channels (CTC) have made
them a serious threat to network security, given the explosive development of data
exfiltration through cyberattacks. Inter-arrival delays are exploited by these kinds of channels
to steal confidential information from the networks they target. Machine learning approaches
are becoming more and more important in CTC detection. These algorithms use statistical
criteria to distinguish between legal (overt) traffic flows and malicious (covert) ones.
However, in order to identify and prevent CTCs and lessen the degradation in service quality
brought on by the detection process, covert channels detection must increase in both
performance and precision given the efforts made by cyberattacks to elude detection and the
expanding column of CTCs. In this study, we describe a novel image-based approach to
completely automatic localization and identification of CTC. Our method is predicated on
the finding that traffic from the hidden channels may be transformed into colorful visuals.
Our solution makes use of this insight to automatically identify and pinpoint the malicious
portion (i.e., packet or set) in a communication flow. Our method lessens the reduction in
quality of service caused by blocking the full traffic flow in which covert channels are found
by identifying the covert portions within the traffic flows. To identify covert traffic, we first
transform traffic flows into colorful pictures and then extract characteristics from the images.

Keywords: Snap Catch, Covert Timing Channels, Machine Learning,Communication

Channels, Automatic Detection

52 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

CHATBOT FOR GOVERNMENT SCHEMES

Dhanalakshmi K(Professor),RC. Hariharan, I. Irastus Jasper, P. Kabilan, R. Gowtham
Department of Computer Science and Engineering
PSNA College of Engineering and Technology,Dindigul, Tamil Nadu

ABSTRACT:
In today's digital age, providing Citizens with accessible and comprehensive
information about Government schemes is crucial for promoting transparency, inclusivity,
and Citizenengagement. This paper proposes the development of a chatbot designed to fulfill
this need by offering Citizen’s information on Government schemes, eligibility criteria, and
engagement opportunities through an intuitive interface and diverse scenarios.The chatbot
will leverage natural language processing (NLP) and Artificial Intelligence (AI) technologies
to engage users in conversational interactions, providingrelevant and accurate information
tailored to individual needs and preferences. Through an intuitiveinterface, users will be able
to easily navigate through various Government schemes, explore eligibility criteria, and
access resources to facilitate their participation in relevant programs.

Diverse scenarios will be incorporated into the chatbot's functionality to address a

wide range of user inquiries and scenarios. These scenarios will include exploring available
schemes, determining eligibility, assisting with the application process, fostering interactive
engagement, handling complex inquiries, and gathering user feedback for continuous
improvement.By offering a user-friendly and interactive platform for accessing Government
scheme information, the proposed chatbot aims to empower Citizens to make informed
decisions about their welfare and avail themselves of the benefits offered by Government
programs. Through this initiative, Governments can enhance transparency, efficiency, and
Citizenparticipationin governance, ultimately contributing to socio-economic development
and societal well-being.

53 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

PROSTHETIC SOLUTION FOR LOWER LIMBAMPUTEES

WITH ANKLE MOVEMENT
Marieswari S, Peer Rabiyathul Femina M, Radhagowri S, Riyala Begum S
Department of Biomedical Engineering, P.S.R Engineering College, Sivakasi.

ABSTRACT :
This project aims to revolutionize the prosthetic solutions available to lower limb
amputees specifically targeting the facilitation of motorcycle riding recognizing the profound
impact of mobility and recreational activities on welfare and quality of life our initiative
focuses on developing prosthetic technologies that seamlessly integrate with the complex
movements involved in operating a motorcycle particularly the nuanced actions of braking
and gear shifting typically controlled by ankle movements through precise alignment of brake
and gear controls with residual ankle movements we strive to create an adaptive and intuitive
riding experience that minimizes the learning curve and maximizes comfort and safety we
recognize the profound impact of mobility and recreational activities on individuals physical
and mental wellbeing and thus our focus lies in developing prosthetic solutions that
seamlessly integrate with the complex movements inherent in motorcycle operation central to
our innovation is the alignment of brake and gear controls with residual ankle movements
replicating the natural convenience of the lost limb this project not only addresses the
functional challenges faced by lower limb amputees but also underscores the importance of
inclusive design in fostering a more accessible and inclusive society.

Keywords: Prosthetic solutions, Lower limb amputees, Motorcycle riding, Brake operation,
Ankle movements, Adaptive experience.

54 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

AI ENHANCED RAILWAY PLATFORM NAVIGATION FOR

BLIND PEOPLE
R.Kavitha1 , K.Mathavi2 , P.Mercy3 , M.Suvetha4 , S.Shenbagavalli5
1,2,3,4UG Students, Department of CSE, JP College of Engineering, Tenkasi, Tamil Nadu
5Asst. Prof., Department of CSE, JP College of Engineering, Tenkasi, Tamil Nadu, India
Corresponding Author: [email protected]

ABSTRACT:
This application is used for an enhancing railway platform identification and
navigation for blind individuals through the integration of artificial intelligence (Al)
techniques within an Android mobile application. The application employs Support Vector
Machine, XGBoost algorithm, and Global Positioning and Navigation System
algorithm.Blind individuals often face significant challenges when navigating unfamiliar
environments, particularly railway platforms, which can be complex and crowded so that this
application was developed.The primary aim of this research is to develop an Android mobile
application that utilizes Al algorithms, to assist blind individuals in identifying and
navigating railway platforms independently.the objective is to create a user-friendly and
effective solution that enhances the mobility and autonomy of blind individuals in navigating
railway environments. The developed application represents a significant step towards
improving the accessibility and autonomy of blind individuals in navigating railway
environments, ultimately enhancing their overall quality of life. Further refinement and
deployment of the application hold the potential to positively impact the lives of blind
individuals worldwide.

Keywords: Blind people,Support vector machine,GPNS,Object identification,Navigation.

55 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

FOODEVEN – FOODORDERINGSYSTEMUSING
MOBILE APPLICATION
Hari Prem.R.C,Hethiskarna.S.R.K , Harinandhavelu.N , Karthikeyan.V
Department of Computer Science and Engineering,
PSNA College of Engineering And Technology
ABSTRACT:

In recent years, food ordering applications have become an integral part of

moderndiningexperiences, offering convenience and efficiency to both consumers and food
service providers. These applications streamline the process of browsing menus, placing
orders, and trackingdeliveries, ultimately enhancing the overall dining experience. With the
advancement oftechnology, such applications have evolved to incorporate sophisticated
features that cater tothediverse needs of users, ranging from personalized recommendations to
real-time order tracking. Existing food ordering systems typically rely on a combination of
frontend and backendtechnologies to manage dynamic datasets and facilitate seamless
communication betweenusersand administrators. Challenges such as slow order processing
times and limited communicationchannels between users and administrators can detract from
the overall satisfactionof bothparties. In response to these challenges, we propose a
sophisticated food ordering applicationtailoredspecifically for college canteens. Developed
using the Flutter framework and Dart programminglanguage for frontend development, with
Firebase serving as the backend infrastructure, oursystem aims to address the unique needs of
college students, college staffs and canteenstaffs. This seamless integration empowers
efficient management of dynamic datasets such as menuitems, user orders, and order statuses,
while also facilitating direct communication channelsbetween user and canteen admin panel.
Our proposed system offers advanced features designed to enhance the dining experience
withinthe college campus. A user-friendly interface allows user to browse menu items,
addthemtotheir cart, and place orders with ease. Upon order placement, user receive a unique
order ID, andtheir order details are stored in the order history section for future reference.
Canteen admincaneffortlessly manage incoming orders, mark them as complete upon
fulfillment and trackorderstatus through the admin dashboard. Through the robust integration
with Firebase, ourapplication fosters a symbiotic relationship between user and canteen
admin, ultimatelyenhancing the overall food ordering experience within the college campus.

56 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

FOOD WASTE PREVENTION APP

G Jayanthi, M Vengadesh Pandian, P Nishandhini, S Simeon
Department of Computer Science and Engineering,PPGInstitute of Technology.

ABSTRACT:

This project is used to manage wastage foods in a useful way. Every day the
people are wasting lots of foods. So we have to reduce that food wastage problem through
online. If anyone has wastage foods they are entering their food quantity details and their
address in that application and then the admin maintain the details of food donator. The
donator can create the account and whenever they are having wastage food they can login and
give request to the admin. And the admin also maintain the buyer (orphanage, poor people,..)
details too. After the admin view the donator request and give the alert message like time to
come and collect the food. And the admin collect foods from donator through their nearby
agent then provide to nearest orphanages or poor people. After receiving the food from the
agent by admin and give alert message to that donator. If the donator needs any detail about
the orphanage with helping thought they can give request to the admin and collect the
orphanage details. This project is food redistribution is an enormously successful social
innovation that tackles food waste and food poverty. The user’s details are maintained
confidential because it maintains a separate account for each user.

57 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

ONLINE RATION SHOPPING PLATFORM

Abish Kumar K, Arjun E, Gowtham S, P.Suganya
Department of Information Technology
M.A.M College of Engineering and Technology (MAMCET)

ABSTRACT:
The Ration Shop cannot able to meet the requirement of the user due to over
population of our country, so the processing speed is low. As a result, there is always crowd
of people in the ration shop. Also there is a chance for the illegal usage of our products in the
regular system, i.e. the materials are robbed by making wrong entries in the register without
the insight of the ration card holder. Due to that large amount of money given by government
gets wasted. Hijacking is the main problem in the user ration card. Hence, we have developed
a smart ration card application for all android mobile users. In this system, each user has a
separate authentication login. The user profile will contain the information about their family
members, the materials which are available and has been received and their price list will also
be displayed. The buyer can block the materials needed and they can request to the admin and
the user will receive a confirmation message. By using this message they can buy their stuff
in their corresponding ration shop. The ration shop admin will upload the details which has
been delivered to the respective user. This platform bridges the gap between consumers and
authorized suppliers or government agencies, streamlining the process of accessing
subsidized ration programs. Users can conveniently browse, select, pay for, and monitor their
orders, ensuring transparency, efficiency, and reliability throughout the shopping journey. By
embracing digitalization, the platform not only enhances user convenience but also promotes
transparency, accountability, and modernization in the distribution of essential goods,
ultimately contributing to a more inclusive and efficient ration distribution ecosystem.

Keywords— ,PDS,FPS

58 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

HEART RESCUE:IOT INTEGRATION FOR ADVANCED CPR

FEEDBACKAND DEFIBRILLATION IN CARDIAC
EMERGENCIES
Vanitha K (Assistant Professor), Gokul Raja T, Kavipriya G, Pradeep P
Department of Medical Electronics, Velalar College of Engineering and Technology Erode.

ABSTRACT:
Cardiopulmonary Resuscitation(CPR)is a lifesaving procedure that can be performed
in a variety of situations when a person's breathing or heartbeat stops, such as a heart attack
or near drowning. At that point, both skilled and unskilled individuals can assist the afflicted
individuals with quick and forceful chest compressions. The force sensitive resistor (FSR),
pressure sensor, and heart rate sensor that are positioned on the chest are employed in this
procedure. Over the FSR, the professional hands will compress on the chest area. Physical
pressure, squeezing, and weight can all be detected by the FSR and pressure sensors. It
functions essentially as a resistor that, in response to pressure, alters its resistance value. A
heart rate sensor will gauge the individual's heart rate. To improve the patient's chances of
survival, we can use a defibrillator as soon as the heart rate drops. With the use of an
electrical shock to the chest, a defibrillator can aid in the regeneration of a heart rhythm that
is recorded in the Internet of Things. The caregivers receive information about the patient and
an alert message via the Internet of Things (IoT). Low-cost and environmentally favorable
are the FSR, pressure sensor, and heart rate sensor. It will effectively assist those who lack
training. But regardless of time, this therapy cannot solve every heart issue.

Keywords– cardiopulmonary resuscitation, force sensitive resistor, heart rate sensor, pressure
sensor, IoT, defibrillator.

59 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

SMART VECHILE SAFETY SYSTEM FOR TWO WHEELER

R.P Suriyasankar^1, A.Dhivakar ^2, C.Muthuguru^3, P.Tamilarasan^4

1-ASSOCIATE PROFESSOR, BIOMEDICAL ENGINEERING,

P.S.R ENGINEERING COLLEGE
2, 3, 4 -UG STUDENTS OF BIOMEDICAL ENGINEERING,
P.S.R ENGINEERING COLLEGE
ABSTRACT:
The latest report of the national crime records Bureau (NCRB), as many as 95,622
people lost their lives due to road accidents in India 2023. The two wheelers have accounted
for maximum fatal road accidents (69,240 deaths), contributing 44.5% of total road
accidental death. Tamil Nadu reported as 8,259 deaths due to two wheeler accidents in 2023
[1].So we are going to design a smart vehicle safety system for two - wheeler driver. If a
person doesn't wear helmet, properly not releasing side stand as well as drunk alcohol, then
the vehicle doesn’t start. Also we have planned to monitor the spinal cord pressure of two
wheeler drivers, also alert the rider about their back pain information [2].This paper focuses
on the above mentioned reasons and provide a solution to ride safety.

60 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

REAL-TIME ROBOTIC AVATAR CONTROL USING FUZZY

GAZE-CLASSIFICATION FOR PEOPLE WITH DISABILITY
Saravanan.S 1 ,Ananda Krishnan.M2 ,Cineghavarthani.B3 ,KamalKumar.S4 ,Suriya.V5
1,2,3,4,5 Department of Computer Science & Engineering
1,2,3,4,5 Sree Sakthi Engineering College,Coimbatore,Tamil Nadu,India

ABSTRACT:
Robotic control with gaze-classification has been an area of interest for quite some
time. In this paper, a novel solution of implementing sensing system with intelligent
visualization has been presented. Such a system can have a variety of applications but can be
of exceptional importance for patients suffering with spinal muscular atrophy (SMA)
disorder, which gradually end with paralyzed state. The presented solution incorporates
Optical Gaze System, within a Virtual reality (VR) headset to control a robotic Avatar
remotely based on the gaze position. The robot is being called as avatar due to its ability to
maneuver and provide audio and video streams to the user wearing the VR headset, thus
acting as an extension of the user him/herself. Hence people with permanent disability due to
degenerative diseases such as SMA can benefit from this design by utilizing the avatar to see
things and places where they cannot go themselves. The psychological impetus can assist
tremendously in their treatment. The Fuzzy Algorithm presented here enables a very fast
image classification method for such near-real-time application. Consequently, the processing
and decision making of the system takes about a second with an accuracy of over 95%. Index
Terms— SMA, Gaze Pattern Matching, Remotely controlled robotic avatar, Adaptive Neuro
Fuzzy Inference System.

61 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

DESIGN & ANALYSIS OF CIRCULAR AND ELLIPTICAL

ANTENNAS FOR UWB APPLICATIONS

P. Venkatesh ( Assitant professor),M.Badri Chelva , L.Jeya Selva Karthickeyan

Department of Electronics & Communication Engineering,
Ramco Institute of Technology, Rajapalayam, India.

ABSTRACT:
This paper presents a thorough investigation into the design and analysis of circular
and elliptical antennas tailored for Ultra-Wideband (UWB) applications. Through
comprehensive optimization of antenna geometry and parameters, including impedance
matching, radiation pattern, and efficiency, the study aims to achieve superior UWB
performance. Utilizing various simulation tools and techniques, the comparative advantages
and limitations of circular and elliptical antenna geometries are explored, with emphasis on
bandwidth, radiation characteristics systems. The proposed antenna provides an excellent
return loss of -33 dB at 3 and 9.44 GHz. The antenna achieves a bandwidth of around 10
GHz. The circular structure is used to achieve this good bandwidth. The gain of the antenna is
2.32 and 5.05 dB which are good gain values. The radiation of the antenna is bidirectional
and omni-directional which are suitable for wireless communications

62 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

AN EXPERIMENTAL STUDY ON MECHANICAL

PROPERTIES OF COUROUPITA GUIANENSIS TREE
DR .T.VARUNKUMAR1, R. PRIYADHARSHINI2, M. RAMPRASANTH3,
S. PALANIVEL4, P. PRAVEEN GOKUL5,
1 - Professor and Head, Department of Mechanical Engineering, P.A. College of Engineering and
Technology, Pollachi, Tamil Nadu, India.
2,3,4,5 - Students Department of Mechanical Engineering, P.A. College of Engineering and
Technology, Pollachi, Tamil Nadu, India.

ABSTRACT:
This research project explores the novel application of the cambium cell layer
extracted from the tree species Couroupita guianensis as a key component in the synthesis of
solid-like substances. By combining the natural cambium cell layer with epoxy resin and
hardener, a new composite material is developed, demonstrating unique properties and
potential applications. The cambium cell layer is extracted, characterised, and its structural
characteristics and content are examined at the outset of the investigation. Afterwards, several
combinations of epoxy resin and hardener are combined with varied proportions of the
cambium cell layer in order to examine the effects on the mechanical and physical
characteristics of the final product. The study looks at the produced composite materials'
mechanical strength, flexibility, and thermal stability. The project's results also pave the way
for the creation of environmentally friendly and sustainable materials, which may find use in
the building, automobile, and aerospace sectors.

63 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

FLOOD MANAGEMENT AND EARLY MONITORING

SYSTEM

Mr.Suresh Kumar .K ,Praveen Kumar . N, Jenith .R , Praveen . A

Department of Electronics and Communication Engineering,
Tagore Institute of Engineering And Technology

ABSTRACT:
The system monitors environmental elements, which include humidity, temperature,
water level, water increase rate, and rainfall, to detect a flood. Then the flood prediction is
done using IOT enabled sensors data with the help of THINGSPEAK APP. Time Series
Forecasting is used in this project to describe an innovative technique to flood prediction. To
analyze, the previously acquired runtime data of a given period of time is taken by the
prediction model in the form of a dataset and flood risk is anticipated.

64 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

IoT BASED INTELLIGENCE ROBOT

Mrs.B.Rama, R. Dhanvarshan, M. Sathish, K.Vishnuprakashraj, S.Kaleel Ahamed
Dept. InformationTechnology(IT)
M.A.M College of EngineeringandTechnology(MAMCET)
(Anna University ofAffiliation)

ABSTRACT:
Worker safety in industries is vital for human well-being and operational
effectiveness. Despite safety advancements, accidents persist due to human error and
hazardous conditions. This paper advocates for Worker Safety Robots (WSRs) to enhance
safety. WSRs integrate robotics, AI, and sensors to detect hazards and assist workers.
Equipped with cameras and gas detectors, they monitor environments in real-time and use
machine learning to predict risks. WSRs collaborate with humans, offering alerts and
navigation in complex areas. Their implementation promises accident reduction, boosting
morale and productivity. WSRs epitomize a collaborative, tech-driven approach to safety,
revolutionizing industrial practices for safer workplaces .
KEYWORDS - Hazard Detection Systems, Emergency Response Robotics, Personal
Protective Equipment (PPE) Monitoring, Collaborative Safety Robots, Real-time Risk
Assessment, Remote Monitoring and Control, Human-Robot Interaction Safety, Predictive
Maintenance Robots, Ergonomic Assistance Robots, Compliance and Regulatory Compliance
Solutions

65 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

AUGMENTED REALITY WITH DYNAMIC RENDERING

FOR EDUCATION
Dr.S.Saravanan1, Kiruba Karan.P2, Kiruba Mohan.P3, Naveen Kannan.S4
Sree Sakthi Engineering College ,Karamadai

ABSTRACT:
Augmented Reality (AR) technology has the potential to revolutionize education by
providinga new way for students to visualize and interact with complex concepts. A system is
proposedto develop an AR smartphone application that allows students to visualize objects
and scenariosthat the teacher is teaching in real-time. Students will be able to view these
models from anyangle and interact with them in a variety of ways, such as by rotating them
or zooming in onspecific details. In addition to enhancing student’s understanding of the
material being taught,the AR application will also provide an engaging and immersive
learning experience. Thedistinguishing factor is the storage of the 3D assets on the cloud that
will equip the educatorwith the option of pre-planning and customizing entire lesson as well
as storing any number ofmodels. It help to increase student engagement and motivation,
leading to better retention ofthe material being taught, the proposed AR smartphone
application has the potential tosignificantly improve the way students learn and understand
complex concepts, makingeducation more effective and enjoyable for all.

Keywords: Augmented Reality, Education, Mobile Devices.

66 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

IDENTIFICATION OF LIVER HEALTH USING MACHINE

LEARNING ALGORITHMS
Subhashini G(Assistant Professor), Rajalakshmi R(Assistant Professor), Ashokkumar
R, Kizhore Kumar SS
Krishna Kumari L
Department of Electronics & Communication Engineering
Ramco Institute of Technology
Rajapalayam, India

ABSTRACT:
Liver disease is one of the biggest global health issues, with millions of cases reported
year. Predicting liver illness early on can improve patient outcomes significantly by allowing
for personalized treatment strategies and timely intervention. In this work, we present a
machine learning-based method for predicting the risk of liver disease. This method uses
machine learning to find patterns in massive amounts of data in order to facilitate decision-
making, enabling machines to learn in supervised, unsupervised, semi-supervised, or
reinforced contexts. Here, decision tree, random forests, KNNeighbor, and logistic
regressions are the algorithms utilized to predict liver patients. To improve model
performance and interpretability, the most useful predictors are found using feature selection
strategies. The analysis and result computations show that the method obtained good
accuracy after picking characteristics.

Keywords—SVM, LSTM, CNN, RF, DT, LR, KNN

67 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

AMAZON RECOMMENDATION SYSTEM

R.Sridevi1, R.Loka Nandan2
1
Final Year, Department Information Technology, KamarajCollege of Engneering and Technology,
Virudhunagar, Tamilnadu
2
Final Year, Department Information Technology, KamarajCollege of Engneering and Technology,
Virudhunagar, Tamilnadu
1
[email protected],[email protected]

ABSTRACT:
There have been growing interests in the area of recommender systems using machine
learning techniques. As there are a great number of explicit and implicit features that can be
used for estimating user preference, it requires scalable and accurate algorithms along with a
system with high availability and scalability. Alternating least square matrix (ALS) algorithm
is an enhanced version of latent factor models using matrix factorization with good scalability
and predictive accuracy. Apache Spark is an open-source distributed platform for processing
big data, achieving good speed and scalability suitable for iterative machine learning
algorithms. Amazon offers cloud computing services with various functionality including
data storage and processing engines and is highly available and scalable. In this study, we
applied the ALS algorithm using Apache Spark running on an Amazon Web Service (AWS)
Elastic Map Reduce (EMR) cluster for recommending a product with a good accuracy and
enhanced scalability.

Keywords: Alternating Least Squares (ALS) algorithm,Latent factor models, Matrix

factorization, Apache Spark, Elastic Map Reduce (EMR)

68 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

A ROBOTIC STEWARD FOR MARINE CLEANUP AND

ECO-SYSTEM SURVEILLANCE
Prof. T. Sathya1 P. Indra2 A.Jeevitha3 M. R. Kasiraj4 K. Saran5
Assistant Professor1, Electronics and Communication Department,
B E Final Year Student 2345, Electronics and Communication Department.

ABSTRACT:
The Automatic Marine Debris Collector and Ecosystem Monitoring Mechanoid is
an innovative solution to the growing problem of marine debris and environmental
degradation. The system is the combination of debris collection with real- time environmental
monitoring, the system can help to protect marine ecosystems and promotes the sustainable
use of ocean resources. In addition to collecting debris, the system also monitors the water
quality, temperature, and other environmental factors. This information is transmitted in real-
time to a remote monitoring station, allowing researchers and environmentalists to analyze
the data and make informed decisions regarding conservation efforts. Developing a WiFi-
based boat with Android application control is an intriguing endeavor that combines
hardware and software skills in the field of robotics and do-ityourself projects. To build a
flexible robot that can be controlled by an Android application created specifically for it
using MIT App Inventor, a number of components must be integrated. The Arduino
NodeMcu, DC gear motors, the L298 motor driver, and several other electrical parts that
support the robot‟s operation are the main parts of this project. This system includes
Temperature and Humidity monitoring system, GAS level monitoring system, Security
system, water level monitoring system, and controlling system. For that, we used the DHT11
sensor, MQ2 sensor, Ultrasonic sensor, and 2 channel relay module. Also, we can monitor
these data on the smartphone, or computer. For that we used the blynk updated app. This app
acts as a user friendly for seamless device cloud communication.

69 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

ONLINE LOCAL FARMER MARKET

Mr.A.Vijay ,1.Nandhitha.R,2Sabarna.T,3Santhiya.P,4Suriya.P
1
Professor, Department of Information Technology,
M.Kumarasamy College of Engineering,Karur,TamilNadu,India-639113
2345
UG Students, Department of Information Technology, M.Kumarasamy College of
Engineering,Karur,TamilNadu,India-639113

ABSTRACT:
The Online Local Farmer's Market is poised to redefine the landscape of agricultural
commerce by introducing a comprehensive digital platform centered around four pivotal
processes. Livestock Sale with Location establishes a transparent and geographically precise
avenue for farmers to showcase and sell their livestock, enhancing accessibility for
consumers seeking locally sourced products. The Vegetable Selling component streamlines
the marketing and sale of produce, providing farmers with a dedicated space to exhibit and
distribute their agricultural goods efficiently. Working Crew Booking introduces a
collaborative dimension, enabling farmers to connect and coordinate with agricultural
workers seamlessly. This facilitates not only the exchange of services but also the cultivation
of a supportive farming community. Simultaneously, the Vehicle Renting process offers a
solution to transportation needs, allowing farmers to easily rent vehicles for the transportation
of goods. Through these interconnected processes, the platform aims to alleviate the
challenges faced by both farmers and consumers, fostering transparency, efficiency, and
collaboration in local agriculture. The vision is not merely to create an online marketplace but
to cultivate a dynamic ecosystem that enhances the accessibility, sustainability, and
profitability of local agricultural practices.

Keywords :Transparency,Streamlined Marketing,Sustainability,Working Crew Booking

70 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

VEHICLE POWER GENERATION SYSTEM USING FOR

STREET LIGHTING
Mrs.Narkees Begam ,Assistant Professor ,ECE,NPR College Engineering &Technology.
Nandhini.A,Malavika.A,V.Nivethitha
Electronics and communication Engineering
NPR College Of Engineering &Technology,Dindigul.

ABSTRACT:
The escalating demand for electrical power in the modern world, coupled with
the dwindling resources of conventional energy sources, underscores the urgent need for
alternative and sustainable energy solutions. This proposed system explores the utilization of
the piezoelectric effect, whereby certain materials can generate an electrical charge in
response to applied pressure and strain. Specifically, the research investigates the application
of piezoelectric technology in the context of auto street lighting for transportation facilities.
By integrating piezo sensors into roads or vehicle tires, the energy produced by the pressure
of moving vehicles is captured and converted into electrical charge using Lead Zirconium
Titanate (PZT) as a Piezoelectric Transducer. This harvested energy is then stored and
utilized as a source of roadside power generation, powering auto street lighting systems. The
effectiveness of this energy generation method hinges on various factors including the system
configuration of PZT, stress applied to the PZT, displacement from its original state, and the
connection method of PZT plates, all of which influence the generated voltage and current.
This research underscores the potential of piezoelectric technology as a sustainable solution
for meeting energy demands and promoting environmentally friendly practices.

Keywords—(Piezoelectric effect,Sustainable energy, Smart infrastructure, Street lighting,

Energy conservation, Automation circuitry.)

71 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

OPTIMIZING DATA MINING RESEARCH: LEVERAGING

WEKA'S BEST FIT TOOL FOR ENHANCED
EFFECTIVENESS
Dr. S. Dinakaran Mrs. Jenil Kiruba
Associate Professor, HCL Tech
T. John College Bangalore
Bangalore-83

ABSTRACT:
Machine learning plays a dynamic role in designing and developingalgorithmsfroma
dataset. Machine learning acts effectively and aggressively in the active fields of medical,
financial, and education sectors, researchers, and online portals with the associated datasets.
Data mining techniques are one of themethods to adopt with machine learning algorithms by
an analytical process. Data analytics is one of the growing and future assumptions for all
research areas because data is one of the most important aspects in all intensities. There are
various simulation tools for effective data mining research. The data must be preprocessed for
further usage and then different techniques might be experienced. The tool tests the datasets
with various data mining algorithms and the working mechanism of several methods can be
visualized. Before researchers step into their research they have an enormous conflict that
which tools to identify so this paper will bring them a clear idea to select the appropriate
tools.

72 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

WATER RESOURCE MONITORING SYSTEM USING

MACHINE LEARNING
V.Kalaiselvi,D.Janet Abiya,M.Reena,S Renugadevi
Department of Computer Science and Engineering,
JP College of Engineering

ABSTRACT:

This study evaluates the effectiveness of Convolutional Neural Networks (CNN) and
Naive Bayes classifiers in mapping water bodies from Sentinel-1 SAR images. Through deep
learning and probabilistic approaches, we aim to identify which methodology provides higher
accuracy and efficiency in distinguishing water surfaces from other land covers. Water body
mapping is crucial for environmental monitoring, disaster management, and urban planning.
Sentinel-1 SAR images offer all-weather, day-and-night observation capabilities, making
them ideal for continuous monitoring. CNNs, with their capacity for feature extraction in
image data, and Naive Bayes classifiers, known for their simplicity and speed, are tested
against this challenging dataset. The models were trained on a diverse set of Sentinel-1 SAR
images, capturing various geographical and climatic conditions. CNNs demonstrated a
significant advantage in accuracy, achieving a 92% success rate in correctly identifying water
bodies, attributed to their ability to process complex spatial relationships in SAR data. Naive
Bayes, while only achieving an accuracy of 70%, proved to be much faster in processing
time, offering a viable option for real-time or resource-constrained applications. The
comparison highlights CNN's superiority in extracting detailed features from SAR images for
water body mapping, offering a robust tool for accurate environmental monitoring. However,
the Naive Bayes classifier's speed and efficiency present a practical alternative for
applications requiring quicker, albeit less precise, analysis. Future work should explore
hybrid models to balance accuracy with computational efficiency.

73 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

THE ADVANCED AUTOMATED BANKING TERMINAL

Mrs.B.Rama, Abder Shaheen A M, Abdul Kader S H, Sancheevi R
Department of Information Technology
M.A.M College of Engineering and Technology (MAMCET)

ABSTRACT:
The Advanced Automated Banking Terminal (AABT) is a sophisticated and
innovative system designed to revolutionize the banking experience through the integration
of advanced technologies. At the heart of this system is the implementation of cutting-edge
biometric authentication methods, interactive voice recognition, and intuitive user interfaces
to ensure both security and convenience for users. One of the key features of the AABT is its
utilization of face recognition technology for user authentication. When a person enters the
ATM, the system immediately initiates the authentication process by scanning and verifying
their facial features. This biometric authentication adds a layer of security by ensuring that
only authorized individuals can access banking services. Moreover, the AABT addresses a
unique challenge posed by twins through its twin detection and fingerprint authentication
mechanism. In cases where twins are identified, the system prompts them to provide their
fingerprints for additional authentication. This extra step enhances security measures and
mitigates the risk of unauthorized access, especially in scenarios where facial features may be
similar. The system's integration of voice recognition technology enables users to interact
with the ATM using voice commands. This hands-free approach to banking transactions
allows users to perform a wide range of functions such as cash deposits, withdrawals, balance
inquiries, and more simply by speaking to the system. The voice recognition system
accurately interprets and processes user inputs, making banking operations more intuitive and
user-friendly. Furthermore, the AABT features an interactive display that provides users with
real-time information and feedback during transactions. Transaction details, account
balances, and confirmation prompts are displayed on the screen, ensuring that users have
clear visibility and control over their banking activities. In terms of 1’st Mrs.B.Rama Dept.
Information Technology (IT) M.A.M College of Engineering and Technology (MAMCET)
(Anna University of Affiliation) Trichy, India [email protected] 2’nd Abder Shaheen A
M Dept. Information Technology (IT) M.A.M College of Engineering and Technology

74 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

(MAMCET) (Anna University of Affiliation) Trichy, India [email protected]

3’rd Abdul Kader S H Dept. Information Technology (IT) M.A.M College of Engineering
and Technology (MAMCET) (Anna University of Affiliation) Trichy, India
[email protected] 4’th Sancheevi R Dept. Information Technology (IT) M.A.M
College of Engineering and Technology (MAMCET) (Anna University of Affiliation)
Trichy, India [email protected] transaction flexibility, the AABT offers users the
choice between withdrawing cash from the ATM's vending machine or depositing funds
directly into their accounts. This flexibility caters to the diverse needs and preferences of
users, allowing them to choose the most convenient option for their banking transactions.
Overall, the Advanced Automated Banking Terminal represents a significant advancement in
banking technology, offering a seamless blend of robust security measures, intuitive voice-
based interactions, informative visual displays, and transaction versatility. By leveraging
these advanced technologies, the AABT aims to enhance the overall banking experience for
users while ensuring the highest standards of security and reliability

75 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

PHISHING MAIL DETECTION USING LARGE LANGUAGE

MODELS
Mrs.C.EYAMINI, Assistant Professor, Department of CSE, Kathir college of engineering

RAGHUL P, Final Year Student, CSE Department, Kathir college of engineering

VIJAY K, Final Year Student, CSE Department, Kathir college of engineering

KARTHIKEYAN J, Final Year Student, CSE Department, Kathir college of engineering

VIGNESH G, Final Year Student, CSE Department, Kathir college of engineering

ABSTRACT:

Phishing attacks pose a significant threat to cybersecurity, targeting

individuals and organizations by impersonating trusted entities to steal sensitive information.
This project presents an innovative approach to phishing mail detection using large language
models (LLMs) and prompt engineering. Leveraging a pre-trained model from the Hugging
Face library, we design prompts that guide the LLM in classifying email content as either
safe or phishing. This strategy eliminates the need for extensive labeled datasets and fine-
tuning, making it a practical solution for real-world applications.To facilitate user interaction
and streamline the detection process, we developed a user interface using Streamlit. This
allows users to input email content and receive immediate classification results, indicating
whether the email is safe or phishing. The model's performance is evaluated on a set of
diverse email samples, demonstrating its accuracy and efficacy in identifying phishing
emails.Our approach provides a cost-effective and efficient solution for phishing mail
detection, offering flexibility and adaptability to emerging threats. By minimizing
dependency on large datasets and providing a user-friendly interface, this project contributes
to enhancing cybersecurity defenses against phishing attacks.

KEYWORDS:LLM, Prompt engineering, phishing mail, HuggingFace, StreamLit.

76 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

EMERGENCY CONNECT-ANDROID APPLICATION

Sowndharya.V, Nandish.C
Department of Biomedical Engineering
Sri Ramakrishna Engineering College, Coimbatore

ABSTRACT:
Mobile applications have become essential tools in the digital world we live in,
especially in emergency situations. This paper is about an innovative Android software
‘EMERGENCY CONNECT’ that quickly gathers essential family details and vitals
information to expedite emergency medical treatments. This app provides healthcare
professionals quick access to patient data, such as allergies, medical history, and emergency
contacts. This app, which primarily aims to increase emergency response efficiency, gives
healthcare professionals the information they need to make earlier decisions, which
eventually improves patient outcomes and strengthens family support in urgent medical
circumstances.

Keywords: Firebase, Android, API, Android Studio,

77 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

SYNTHESIS OF ANATASE GRADE TIO2 BY OXIDATION

CUM LEACHING PROCESS FROM TITANIUM BEARING
MINERALS
S. VIGNESH
Bachelor of Technology in Chemical Engineering
Kalasalingam Adamic Research And Education

ABSTRACT:
Titanium and its compounds find extensive use in specialized industries including
plastic, paint, and aircraft. make titanium metal and its derivatives are the titanium-containing
minerals rutile (TiO2) and ilmenite (FeTiO3). The present work aims to employ the tubular
furnace oxidation followed by acid leaching to process titania slag from industry to
synthesise anatase grade pure titanium dioxide. The two different percentage titanium
containing slag are used for preparing titanium dioxide. The chemical and physical
characterizations of the raw material and extracted titanium dioxide were done using X-ray
diffraction (XRD), (SEM) and Fourier-transform infrared (FT-IR) spectroscopic analysis.
Results indicated that the tubular furnace roasting has significantly influenced the leaching
and removal of iron impurities to maximum. TiO2 synthesised with the recommended
roasting cum leaching approach highlightsthat a potential resource of titanium for the
manufacture of high-quality anatase grade TiO2 applicable in industries.

Keywords; Rutile, anatase, ilmenite, tubular furnace; roasting

78 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

ENHANCING MICROBIAL FUEL CELL EFFICIENCY BY

USING EGGSHELL POWDER WITH SACCHAROMYCES
CEREVISIAE IN BATCH PROCESSES
R RAGHURAM SHANKAR
Department of Chemical Engineering
Kalasalingam Academy Of Research And Education

ABSTRACT:
Microbial fuel cells (MFCs) present a promising avenue for sustainable energy
production by harnessing the metabolic activities of microorganisms. This study explores the
feasibility of employing eggshell powder (ESP) and Saccharomyces cerevisiae, a commonly
used yeast strain, as substrates in an MFC for electricity generation. ESP, rich in calcium
carbonate, serves as an inexpensive and eco-friendly alternative to traditional catalysts, while
S. cerevisiae offers efficient electron transfer capabilities.The MFC setup comprises an anode
chamber inoculated with S. cerevisiae and ESP as the substrate, and a cathode chamber
containing an oxidizing agent. Through microbial metabolism, S. cerevisiae oxidizes organic
compounds present in ESP, releasing electrons that travel through an external circuit to the
cathode, where they combine with the oxidizing agent, generating electricity. This abstract
summarizes the experimental approach, including the optimization of MFC parameters such
as electrode material, pH, and temperature, to enhance electricity generation. Additionally,
characterization techniques such as cyclic voltammetry and microbial community analysis are
employed to evaluate the performance and microbial diversity of the MFC system.
Preliminary results demonstrate the feasibility of utilizing ESP and S. cerevisiae in MFCs for
electricity generation, with observed improvements in power output and microbial activity
compared to traditional substrates. This research contributes to the development of
sustainable energy technologies by utilizing readily available waste materials and microbial
resources for electricity generation.

79 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

DETECTING AND CLASSIFYINGMALWARE WITH

MACHINE LEARNING
Jebin Bose S Kalaiselvi R
Research Scholar/ Associate Professor/
Department of Computer Science and Engineering Department of Computer Science and Engineering
Noorul Islam Centre for Higher Education, R.M.K. College of Engineering and Technology,
Kumaracoil. Puduvoyal,
Thuckalay, Tamil Nadu, India Gummidipoondi
[email protected] [email protected]

ABSTRACT:
With the rapid evolution of the web, malware has emerged as a significant digital threat in
contemporary times. Consequently, the detection of malware stands as a crucial element in
safeguarding computer systems. Presently, attackers commonly deploy polymorphic malware, a
type of malicious software that continuously alters its identifiable traits to evade conventional
signature-based detection methods. Hence, the necessity for Machine Learning-based detection
becomes imperative. This study aims to extract behavioral patterns, achievable through static or
dynamic analysis, followed by the application of diverse ML techniques to discern whether the
software is malicious or benign. Emphasis will be placed on behavioral-based detection
methods, leveraging ML algorithms to formulate a robust model for social-based malware
recognition and classification.

Keywords—Machine Learning in the Context of Malware Detection: Exploring KNN

and SVM.

80 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

ALEXA HOME AUTOMATION SYSTEM USING NODEMCU

ESP8266
R.Rasika, S.Prakash Karthik, K.Pavendhan

Department of Electrical and Electronics Engineering,

Dr.Mahalingam College of Engineeerimg and Technology, Pollachi.

ABSTRACT:
Voice- Controlled automation of domestic chores and ind increased convenience may
be achieved with the NodeMCU ESP8266 Alexa Home Automation System, which is a
creative solution. Users may easily manage and monitor a variety of household appliances
and gadgets by utilising the capabilities of NodeMCU ESP8266 microcontroller board and
merging it with Amazon Alexa’s speech recognition and control functionalities. Users may
operate lights, fans, switches, and other network-connected smart devices with voice
commands thanks to the system’s design, which enables smooth communication between the
NodeMCU board and the Alexa Voice Service. Additionally, the system’s scalability makes
it simple to expand accommodate more smart devices, and strong security measures
guarantee that user privacy and data are protected. The Alexa Home Automation System with
NodeMCU ESP8266 enables customers to construct personalised.

81 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

MODELLING OF EV USING SIMULINK

Tharani Prakash N, Kanisha A D , Jeeva Prasath M R
Department of Electrical and Electronics Engineering,
Dr Mahalingam College of Engineering and Technology, Pollachi.

ABSTRACT:
We present a Simulink model for electric vehicles (EVs). We model the vehicle
elements, longitude driver, Driver cycle,H-Bridge,PM-DC motor and battery of the EVs to
gain the power requirements of the battery and to later derive the best kinds of battery to use
for such applications. The reenactments are performed through a joining of the Simulink
blocks. The speed of the vehicle and its distance gone compare to real driving cycles and
force varieties of an EV. The additional worth of this model is that it reenacts the streamlined
drag, SOC, straight speed increase, and moving opposition powers where the demonstrated
engine productivity hit 73%. Different outcomes comparing to its tractive impact, engine
proficiency, and battery necessities are gotten. The reenactments are confirmed and all fit the
hypothesis behind Evs. Keywords: Electric, vehicle, simulation, H-Bridge , SOC,PM-DC
motor.

82 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

ASMART OBSTACLE DETECTOR AND PERSONAL VOICE

ASSISTANT FOR BLIND PEOPLE USING IOT
Jini Kumari.M, Nithya.V
Mechatronics,
T.S Srinivasan Centre for Polytechnic CollegeAdvanced Trainingand Advanced Training,
Chennai

ABSTRACT:
Visual Disability is one of the global issue. They face several challenges and find
themselves very difficult to self navigate in known and unknown environment.An intelligent
smart personal voice assistance for visually impaired and blind people using IOT is a
technology-based solution designed to assist blind and visually impaired peoples(BVIPs) in
navigating their surroundings. This system offers a fresh kind of assistive technology for
persons who are blind or visually impaired. The proposed system is a wearable helmet that
utilizes image recognition, voice generation modules cameras to detect and analyse the
environment, providing real-time audio and tactile feedback to the user. The system aims to
enhance the user’s mobility, independence, and safety by detecting obstacles, providing
directional guidance, identifying objects, and facilitating communication with the
environment. This blind assistance system represents a significant breakthrough in improving
the quality of life for visually impaired individuals and promoting inclusivity and
accessibility. They can walk alone, they do not need anyone to assist them, they can handle
their self correctly with this solution. This system encourages persons who are blind or have
vision impairments to be independent and self-reliant.

Key Words:Real time audio, Internet of things, Blind assistance,BVIPs-Blind and visually impaired
peoples.

83 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

HEALTHCARE SYSTEM USING EMOTIONAL

INTELLIGENCE
L.Shangeetha, CSE [email protected]
V.Dharshini, CSE [email protected]
Guided By Dr. S.Padmapriya, Head of department, CSE
A.V.C College of Engineering , Mannampandal, Mayiladuthurai.

ABSTRACT:
The integration of emotional intelligence (EI) into healthcare systems marks a
transformative approach towards enhancing patient care and overall well-being. This paper
explores the significance of EI in healthcare settings, focusing on its potential to revolutionize
patient-provider interactions, improve treatment outcomes, and foster a more empathetic
healthcare environment. By incorporating EI principles into clinical practice, healthcare
professionals can develop a deeper understanding of patients' emotional needs, leading to
more personalized and effective care strategies. Moreover, the application of EI extends
beyond patient care, influencing organizational dynamics, team collaboration, and leadership
within healthcare institutions. This abstract highlight the key components of EI integration in
healthcare, including empathetic communication, self-awareness, social skills, and emotional
regulation. Through case studies and empirical evidence, the paper underscores the tangible
benefits of adopting an EI-centric approach, such as enhanced patient satisfaction, reduced
healthcare disparities, and improved provider wellbeing. Ultimately, the implementation of
EI in healthcare systems holds promise for cultivating a culture of compassion, resilience,
and patientcentered care, thereby advancing the overarching goal of promoting health and
healing in diverse communities.

84 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

MICROWAVE SENSING OF QUALITY ATTRIBUTES OF AGRICULTURAL AND

FOOD PRODUCTS
Mrs.D. JAISAKTHI,BARATH.D, GOKULAKRISHNAN.M, VIMAL.R
Department of Electronics and Communication Engineering
Tagore Institute of Engineering And Technology

ABSTRACT:
India, with more than 1.27 billion population, ensuring food security is very important.
Grain production has been increasing steadily due to advanced production technology but
improper infrastructure facilities and unfavorable environmental conditions result in storage
losses of food grains. The environmental factors like temperature, moisture content,
humidity, and light greatly influence the storage of food grains. Also, the factors like time
and purpose of storage, type of storage, preventive insecticide treatments and storage
practices account for the food storage losses. During storage, both qualitative and
quantitative losses occur due to insect pests, mold growth, rodents, rats, fungi, micro-
organisms and subsequent production of mycotoxins in storage. The occurrence and number
of stored food insect pests are directly related to the climatic and geographical conditions.
The multiplication rates of the almost all the species of insects is very high and may destroy
up to 10-15% of food grains and deteriorate the rest with undesirable odor and flavors.

85 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

DETECTION OF CORONA VIRUS USING LUNG X-RAY

IMAGES USING MACHINE LEARNING
Author 1 –Christina Ranjitham M, Research scholar, Mepco Schlenk Engineering college,
sivakasi.
Author 2 –Gabrial Gnanasekar D, PG scholar, Francis Xavier Engineering College,
Tirunelveli.
Author 3 –Dr. J. Angela Jenifa Sujana, Assoc/Prof, HOD/AI, Mepco Schlenk Engineering
college, sivakasi.

ABSTRACT:
Coronavirus (Covid sickness 2019) is an irresistible illness brought about by serious
intense respiratory disorder Covid 2 (SARS-CoV-2), which is a kind of Covid. The infection
was formally reported as a pandemic by the World Wellbeing Organization(WHO) on 11
Walk 2020. Given spikes in new Coronavirus cases and the re-opening of everyday exercises
all over the planet, the interest for checking the pandemic is to be more underlined. Clinical
pictures and man-made consciousness (artificial intelligence) have been seen as valuable for
quick evaluation to give treatment of Coronavirus tainted patients. The PCR test might
require a few hours to open up, data uncovered from the chest X-beam assumes a significant
part for a fast clinical evaluation. This implies assuming the clinical condition and the chest
X-beam are typical, the patient is sent home while anticipating the consequences of the
etiological test. Yet, assuming that the X-beam shows obsessive discoveries, the thought
patient will be owned up to the emergency clinic for close checking. Chest X-beam
information have been viewed as exceptionally encouraging for evaluating Coronavirus
patients, particularly for settling crisis division and earnest consideration community
overcapacity. Profound learning (DL) strategies in man-made reasoning (simulated
intelligence) assume a prevailing part as elite exhibition classifiers in the identification of the
sickness utilizing chest X-beams. Perhaps of the greatest test following the Coronavirus
pandemic is the recognition of the illness in patients. To address this challenge we have been
utilizing the Profound Learning Calculation to fabricate a picture acknowledgment model that
can distinguish the presence of Coronavirus from a X-Beam or CT-Sweep picture of a
patient's lungs. Move learning has become perhaps of the most widely recognized procedure
that has accomplished better execution in numerous areas, particularly in clinical picture
examination and characterization. We utilized Move Learning procedures like Origin
V3,Resnet50, Xception V3 that are all the more broadly utilized as an exchange learning
strategy in clinical picture examination and they are exceptionally powerful.

86 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

LIBRARIAN’S DELIGHT
Mrs.N.Srija,1Santhiya.K,2Subha Sree.C,3Swathika Raksha.S.B,4Thirisha.N
1
Professor, Department of Information Technology, M.Kumarasamy College of
Engineering,Karur,TamilNadu,India-639113

2345
UG Students, Department of Information Technology, M.Kumarasamy
College of Engineering,Karur,TamilNadu,India-639113

ABSTRACT:
Librarian’s Delight is a system which maintains the information about the books
present in the library, their authors, the members of library to whom books are issued, library
staff and all. This is very difficult to organize manually. Maintenance of all this information
manually is a very complex task. Owing to the advancement of technology, organization of
an Online Library becomes much simple. The Online Library Management has been designed
to computerize and automate the operations performed over the information about the
members, book issues and returns and all other operations. This computerization of library
helps in many instances of its maintenances. It reduces the workload of management as most
of the manual work done is reduced. The daily manual process of the “Libraries” like Book
entry details, Students and Staffs entry details, Book purchase entry details, Book transaction,
and Fine Days Details etc. had been implemented which can be followed through the
computer technology i.e., “Librarian’s Delight”.

Keywords: Book Registration, Book Availability, Book Details, Send Mail

87 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

NON-INVASIVE BLOOD PRESSURE MONITORING USING

PHOTOPLETHYSMOGRAPHY SIGNALS: A
COMPREHENSIVE INVESTIGATION
Sharath S,Archana S
Department of Biomedical Engineering,Sri Ramakrishna Engineering College

ABSTRACT:
Blood Pressure (BP) serves as a crucial indicator of cardiovascular health, reflecting the
pressure exerted by blood against arterial walls. Photoplethysmography (PPG) emerges as a
non-invasive optical method for monitoring changes in blood volume per heartbeat, offering
insights into heart activity. An individual's blood pressure (BP) needs to be closely monitored
in order to identify hypertension in its early stages. Traditional cuff based non-invasive BP
monitoring methods are prevalent, but their bulkiness and obtrusiveness has paved a way for
exploration of cuffless andcontinuous BP measurement technology. Recent innovations in
cuffless technology offer promising avenues for accurate BP estimation without cuffs, though
they often require additional measurements such as electrocardiogram and
photoplethysmogram, as well as periodic calibration using a cuff sphygmomanometer.This
study delves into the potential correlation between PPG signals and various BP parameters,
including Systolic Blood Pressure (SBP) and Diastolic Blood Pressure (DBP). Through
experimentation, observation, and analysis this study aims to uncover relationships between
PPG signals and BP parameters employing a real time method for the SBP and DBP
estimation and improving accuracy through consensus.

Early detection of hypertension is paramount for effective cardiovascular health

management. Thus, continuous and careful monitoring of BP is essential. This study
endeavors to contribute to the development of non-invasive and continuous BP monitoring
methods by examining the potential correlation between PPG signals and BP parameters,
ultimately aiming to enhance early detection and intervention strategies for improved
cardiovascular health outcomes.

Keywords—Blood pressure (BP), Photoplethysmography (PPG), continuous monitoring,

cuffless technology, early detection

88 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

INTRUSION DETECTION WHILE ACCESSING LOG DATA

FILES IN CLOUD DATALAKE ARCHITECTURE
Veera Viswas B, Kaseviswanathan R, Prasshanth S K, Ilakkiyan S
Kathir college of Engineering

ABSTRACT:
Software-Defined Networks (SDN) offer enhanced network management and control,
but also introduce new security vulnerabilities. This paper presents a comprehensive
approach for intrusion detection in SDN environments, combining Elliptic Curve
Cryptography (ECC) for secure data transmission with a hybrid machine learning model for
accurate attack classification. The system utilizes the Curve25519-Dalek-Hash (CDH) key
exchange protocol to encrypt sensitive network data, ensuring confidentiality and integrity. A
hybrid model integrating XGBoost and LightGBM algorithms is employed for efficient and
accurate attack detection. The proposed system is evaluated on a real-world SDN dataset,
demonstrating high accuracy and efficiency in identifying malicious activities.

Keywords - ,Data Lake,Software-Defined Networks, Intrusion Detection, Machine Learning,

Elliptic Curve Cryptography, XGBoost, LightGBM, CDH

89 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

SUSTAINABLE DEVELOPMENT THROUGH POVERTY

ANALYSIS USING RANDOM FOREST ALGORITHM
Dr. Prasath N(Assistant Professor ),Gokul M K, AdhinJibil X
Department of NWC,School of Computing,
SRM Institute of Science And Technology

ABSTRACT:
This abstract outline a comprehensive approach to poverty analysis leveraging
machine learning and data mining techniques in conjunction with census data and
development indices within a country. It involves the utilization of advanced computational
methods to extract meaningful insights from large-scale datasets, incorporating demographic
information from census records and socio-economic indicators from development
indices.Incorporating machine learning algorithms enables the discovery of essential patterns,
correlations, and predictive models, fostering a more profound comprehension of poverty
dynamics and its multifaceted determinants. Through this approach, it becomes possible to
not only assess the current state of poverty but also forecast and strategize interventions for
sustainable development. The synergy between data mining principles, census data, and
development indices presents an innovative framework for policymakers, researchers, and
social organizations to comprehensively analyze and address poverty-related challenges
within a country's socio-economic landscape.

Keywords--- Poverty analysis, census, GDP, public-funding, CSR value, company,

eradication, societal, better-life-style,pure water, BPL, health, nutrition.

90 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

SMART KITCHEN USING INTERNET OF THINKS

S.Vinothini1,Bathala Lakshmi Prathap2, Lokiya P 3, Mathesh B 4,Pavithra M 5
1,2,3,4,5 Department of Computer Science & Engineering
1,2,3,4,5SreeSakthi Engineering College, Coimbatore, Tamil Nadu, India

ABSTRACT:
Although much of the work has been done until today to realize the Internet of Things
(IoT) into practice, most of the work focuses on resource constrained nodes, rather than
linking the existing embedded systems to the IoT network. The Internet of things (IoTs) is a
network of physical objects or things embedded with electronic, software, sensors and
connectivity to enable objects to exchange data with manufacturers, operators and connected
devices. It can be described as connecting everyday objects like smart phones, sensors and
actuators to the Internet where the devices are intelligently linked together enabling new
forms of communication between things and people, and between things themselves. IoT
allows objects to be sensed and controlled remotely across existing network infrastructure. It
also provides efficiency, accuracy, comfort and economic benefit. Now anyone, from
anytime and anywhere can have connectivity for anything and it is expected that these
connections will extend and create an entirely advanced dynamic network of IoTs. Our work
here tries to enhance the Internet oriented approach with semantic oriented methods, both of
which are required to build practical, complex IoT applications, which are expected on rich
embedded devices.

Keywords:Embedded Systems, Manufacturers,Infrastructure,Connectivity, Dynamic

Network,IoT

91 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

AN IOT BASED MEDI-KIT FOR PASSENGERS IN TRAIN

SURENDER SM.E.,( ASSISTANT PROFESSOR), CHARUMATHI P, KALAIMATHI G,
SUGANTHI S
Department of Medical Electronics
Velalar College of Engineering and Technology, Thindal,Erode.

ABSTRACT:
One of the most popular, affordable, and quick modes of transportation for long-
distance travel and the movement of large commodities is the railway. Among the biggest
rails is the Indian Railways. In India, trains are the primary mode of transportation for both
short and long distances. There is no first aid system on trains to assist people who require
medical attention right away. The use of an Internet of Things-based Medi-kit for train
passengers is a major step forward in improving rail travel security and safety. The
project's goal is to reduce health hazards and enhance emergency response capabilities by
utilizing IoT technologies, which will ultimately make travel for passengers safer and more
comfortable.

Keywords – Internet of Things , Wireless Sensor Network , Health Care , Immediate aid,
Efficient.

92 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

OPTICAL SPECTROSCOPY METHOD FOR DETERMINING

AND ANALYSING BLOOD GLUCOSE LEVEL

Aakash Vaishnav.P, Gokul.K, Kishore.V

Department of Electronics and Communication Engineering, Dhirajlal Gandhi College of

Technology, Salem, Tamilnadu

ABSTRACT:

Accurate and frequent glucose monitoring is crucial for individuals with diabetes to
effectively manage their condition. By providing a non-invasive and continuous monitoring
solution, optical-based glucose monitoring sensors offer a significant improvement in the
quality of life for people with diabetes. This technology enables better glucose control,
leading to improved health outcomes and reduced risk of complications. Non-invasive
glucose monitoring is a promising technology that has the potential to revolutionize the way
we manage diabetes. However, there are still a number of challenges that need to be
addressed before non-invasive glucose monitors can become a reality. In this paper, we
propose a non-invasive glucose monitoring system that uses an ESP32 microcontroller, an
Optically equipped ADC, and an OLED display. The system uses the Optically equipped
ADC to measure the optical absorption analysis of a fingertip. This optical analysis is used to
calculate the glucose level followed by calibration based on pre-programmed algorithms. The
results of the glucose measurement are displayed on the OLED display. The system is
designed to be low-cost, portable, and easy to use. The system has a number of advantages
over traditional invasive glucose monitoring systems. It is non-invasive, which means that it
is more comfortable for users.

93 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

DETECTION OF OVARIAN CANCER BASED ON CLINICAL

DATA USING MACHINE LEARNING APPROACHES
Mr.M.Mohankumar, Aswin.J, Dhanalakshmi.I, Kishore.B
Department of Artificial Intelligence and Data Science
KGiSL Institute of Technology Coimbatore ,

ABSTRACT:

Cancer that originates in or on the ovary is ovarian cancer and this cancer has severely
affected aged woman. From many investigations, it is found that ovarian cancer is the fifth
common disease and seventh cause of woman death across the world. For classification of
ovarian cancer, numerous researches work developed methods based on Deep Convolutional
Neural Network (DCNN). The most significant factor considered by doctors to take decisions
is the classification accuracy. If the classification accuracy is high, doctor can take precise
decision to give proper treatment. Death rate can be reduced and lives can be saved if the
disease is diagnosed at accurately and at the early stages. This Project the gynecological
ovarian cancer prediction techniques can be used by Artificial Intelligence technology with
Machine Learning as well as prediction of cancer patient’s potential identity

Keywords: natural language processing(NLP), legal , LLM , fine-tuning.

94 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

EXPLORING REACT-NATIVE WITH NLP: A CASE STUDY

IN CONTENT RECOMMENDATION SYSTEMS
S.M. Hemalatha, S. Bharath Priyan, S. Sarathi, P. S. Varshiga
Department of Artificial Intelligence M. Kumarasamy College of Engineering ,Karur.

ABSTRACT:

The purpose of this paper is to recommend the recommendation of the suggestion of

the individual with the fidelity of increasing the variability of the implicit developmental
process with the natural language process. React-native in the development of mobile apps
Applying its versatility and NLP capabilities to textual data understanding and analysis, this
study explores the feasibility and effectiveness of integrating this technology to develop
contextual recommendation systems using machine learning for content analysis and user
interaction data for personalization , Designed and implemented prototype application
Through testing and evaluation, this paper investigates the performance, usability, and
scalability of the developed system. The findings help provide insights into the potential of
React Native and NLP collaboration to enhance content recommendation systems, thus
providing valuable implications for future research and practical applications in mobile
development and content providing user-centered.

95 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

A DYNAMIC POWER EFFICIENT CMOS COMPARATOR

S.Syed Jalal and S.vigneshkumar

Department of Electrical and Electronics Engineering,

Dr.Mahalingam College of Engineeerimg and Technology, Pollachi.

ABSTRACT:

In order to create a high-speed comparator, this project suggests using a mid-stage

latch circuit. The offset and low kickback noise of the suggested circuit are its benefits.
Furthermore, by utilizing a different stage in place of a direct link between the pre-amplifier
and latch stages, low-power and high-speed characteristics are achieved.We introduce a
three-phase dynamic comparator that uses a delayed clock to break the trade-off between
speed and power. Despite using the delayed clock to increase the pre-amplifier’s gain, which
in turn reduces the latch stage’s delay and enhances the offset voltage, the circuit has
considerable kickback noise and common-mode voltage dependency.We introduce a three-
phase dynamic comparator that uses a delayed clock to break the trade-off between speed and
power. Despite using the delayed clock to increase the pre-amplifier’s gain, which in turn
reduces the latch stage’s delay and enhances the offset voltage, the circuit has considerable
kickback noise and common-mode voltage dependency.

96 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

CULTIVATING AGRICULTURAL INSIGHTS: HARNESSING

HYPERSPECTRAL DATA FOR BIG DATA ANALYTICS
Dr.M.Ramakrishnan 1 , Mr.Y.Harishroshan 2 , Mr.G.Krithik Balaji 3 , Ms.P.Shalini 4 ,
Ms.S.Tharchana 5
Assistant Professor 1 , Students 2 3 4 5 , Department of Information Technology, Sengunthar
Engineering college, Perundurai, Erode, Tamil Nadu, India.

ABSTRACT:
Hyper spectral and multispectral information processing systems and technologies
have demonstrated its usefulness for the improvement of agricultural productivity and
practices by providing useful information to farmers and crop managers on the factors
affecting crop status and growth. These technologies are widely used in a range of agriculture
applications such as crop management, crop yield forecasting, crop disease detection, and the
monitoring of agriculture land usage, water, and soil conditions. Hyperspectral information
sensing can acquire several hundred spectral bands that cover the electromagnetic spectrum
of an observational scene in a single acquisition. The resulting hyperspectral data cube
contains a large volume of spatial and spectral information. In this system, hyper-spectral
images are processed and classified using a combination of the 2D-DWT algorithm for image
data preprocessing and the Extreme Learning Machine (ELM) algorithm for image
classification. The process involves extracting pixel information, applying 2D-DWT to
enhance feature representation, and utilizing ELM for classification. ELM stands out for its
rapid training and its ability to work with hidden nodes whose parameters are randomly
assigned and not iteratively updated, making it highly efficient for classification tasks. The
system significantly improves hyper-spectral image classification, providing a streamlined
and effective solution for analyzing and categorizing these complex data sets.

KEYWORDS: Hyperspectral image, Crop yield forecasting , Crop production,ELM.

97 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

VISUAL FIRE ALARM SYSTEM: LEVERAGING OPENCV

FOR DETECTION
Yazhini S(Assistant Professor), Kayalvizhi R, Sowmiya S, Subashini S

Information Technology, Nandha Engineering College-Erode

ABSTRACT:
The suggested system combines Open CV with real-time video feeds to detect fire
alarms in real time. Pre-processing involves converting frames to the HSV colour space and
using thresholding to separate out areas that resemble flames. Morphological techniques
improve accuracy and reduce noise by further refining the segmented areas. Fire alarms are
located by looking for contours in these areas. With possible applications in fire monitoring
and safety systems, this methodical technique seeks to offer a dependable way for detecting
fire alarms. The technology can effectively analyse video feeds and enable early detection
and response to fire situations in a variety of environments by utilising computer vision
algorithms on live camera streams. Its capacity to recognise particular visual clues connected
to fire alarms makes it beneficial in enhancing safety protocols and maybe saving lives. Tight
testing and optimisation, however, will be necessary to guarantee the system's dependability
and functionality in a variety of settings.

Keywords--Fire Detection, Alarm System, Open Cv, Emergency Response

98 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

BLOCK CHAIN WASTE MANAGEMENT USING SECURE

DATA STANDARD
Ragunath.R(Assistant Professor), Kaviyadharshini. K, Kayalvizhi.P, Meenakshi. S
Information Technology, Nandha Engineering College-Erode

ABSTRACT:

Old devices are abandoned and replaced. E-waste is the word used to describe these
Electronic and Electrical Equipment’s (EEEs) that consumers abandon. E-waste management
involves collecting, sorting, and recycling old electronics in an environmentally responsible
way. The unorganized sector still controls a substantial portion of India's e-waste collection
and recycling industry, which makes it exceedingly challenging to implement such rules.
Using the DES (Data Encryption Standard Algorithm) and block chain technology, we
present a novel solution for Modified EWM. Block chain is the technology that allows us to
build smart contracts. Smart cities can prevent environmental issues brought on by
inappropriate garbage disposal by lowering air pollution, safeguarding the aquatic ecology,
and enhancing public health. Nonetheless, the technologies, methods, and systems used for
waste management today remain centralized and manual. They are therefore susceptible to
manipulation and the issue of the single point of failure. Furthermore, a lot of the current
waste management systems in smart cities don't offer characteristics like audit, security,
traceability, operational transparency, and trustworthy data provenance.

Keywords: Waste Management, Blockchain, Traceability, and Smart Cities

99 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

BIOELECTRICITY GENERATION FROM BANANA PEELS

WASTE USING DOUBLE CHAMBER MICROBIAL
FUELCELL
Niraj Rambhau Sumatkar,Niraj Sumatkar
Department Of Chemical Engineering,Kalasalingam University

ABSTRACT:
This study investigates the feasibility and performance of microbial fuel cells
(MFCs) utilizing banana peel waste as a sustainable substrate for electricity generation.
Banana peels, abundant agricultural byproducts rich in organic matter, present a promising
alternative to synthetic substrates in MFCs. The research focuses on optimizing MFC
configurations and operational parameters to maximize power output and microbial activity
using banana peel extract as the sole carbon source. Experimental results demonstrate the
capability of banana peel-based MFCs to efficiently convert organic waste into electrical
energy, with comparable or even superior performance to conventional MFC setups.
Furthermore, the study explores the influence of various factors, such as electrode materials,
microbial inoculum, and substrate concentration, on MFC performance and stability. Overall,
this research contributes to the advancement of sustainable energy technologies by
showcasing the potential of banana peel waste as a renewable resource for electricity
generation in microbial fuel cells.

100 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

MINDMATE CHATBOT: AN AI-DRIVEN APPROACH TO

MENTAL HEALTH CARE
Karthickthasan S, Aakash R, Aarthy R, Abinaya S, Suriya A
Department of Computer Science and Engineering , KGiSL Institute of Technology
{karthickthasan.s, ashaakash219, raarthy1553, abinayasivakumar20,
suriyasri92}@gmail.com
ABSTRACT:

Mental health issues are increasingly prevalent worldwide, affecting millions with
conditions such as stress, loneliness, and anxiety. Traditional approaches to mental health
care often face challenges in accessibility and providing personalized treatment. In response,
we introduce MindMate, an innovative digital platform designed to enhance mental health
support through the integration of advanced technologies. At its core is Eva, a generative AI
chatbot that engages users with real-time companionship and support, developed using a
specialized mental health dataset. The platform also includes a comprehensive blog with
articles and resources on mental health, alongside a detailed assessment test designed to
evaluate stress levels effectively. MindMate offers distinct modules tailored for patients,
doctors, and administrators. Patients have facilities to locate healthcare providers, manage
appointments, and access various services seamlessly. For doctors, the platform provides
tools that streamline patient management and enhance communication. The administrative
module is tasked with overseeing essential operations such as user and doctor registration and
affiliating hospitals. MindMate aims to make mental health care more accessible and
personalized, addressing the limitations of existing services and providing scalable, effective
support for individuals seeking help.

Keywords: Artificial Intelligence, Machine Learning, Large Langauge Models, AI ChatBot,

Mental Health Chatbot, Generative AI

101 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

DYNAMIC SCHEDULING IN EDGE-CLOUD COMPUTING

ENVIRONMENTS USINGSEAGULL OPTIMISATION
ALGORITHM
S. Supriya
Research Scholar, Department of Computer Science,
Kongunadu Arts and Science College
Coimbatore - 641029.
Email: [email protected]

Dr. K. Dhanalakshmi
Head of the Department, Department of Information Technology,
Kongunadu Arts and Science College,
Coimbatore – 641029.
Email: [email protected]
ABSTRACT:
Internet of Things (IoT) device counts have increased phenomenally as a result of
technology’s widespread adoption and integration. Consequently, enormous amounts of data
have been generated that must be transferred, processed and stored in order to be interpreted
and used effectively. The final goals of scheduling are to maximize performance, reduce
latency and improve experience quality. Because of restricted behaviors, it is difficult to plan
application jobs efficiently in these situations. This makes scheduling and resource
monitoring even more challenging. In an effort to lower metrics such as SLAV (Service
Level Agreement Violations), AMT (Average Migration Time), ART (Average Response
Time), and AEC (Average Energy Consumption)Seagull Optimization Algorithm is
proposed. The recommended approach minimize number of iterations while achieving the
trade-off between waiting time, cost, load balancing and energy usages. The experiment
findings demonstrate that the suggested technique outperforms a variety of meta-heuristic
algorithms, including Genetic Algorithm (GA), Particle Swarm Optimization (PSO), Ant
Colony Optimization (ACO) and Whale Optimization Algorithm (WOA) in terms of work
scheduling. Second, Multi-Layered Regression Host Employment (MLRHE) clustering is
introduced for detecting VM loads. This paper proposes a time series forecasting model
called Long Short-Term Memory (LSTM). LSTM model is suggested in a smart logistics
framework to improve task scheduling and resource allocation.This study examines a few of
the most significant scheduling strategies and makes the hypothesis that accurate approaches
may be possible through deep learning techniques. The energy consumption, response time,
SLAV and operational cost of the experiment are all measured.

Keywords: Task Scheduling, Seagull Optimization Algorithm, Long Short-Term Memory,

Edge computing, Cloud Computing.

102 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

LEAF DISEASE DETECTION USING DEEP LEARNING

R JAYAPRIYA
Assistant Professor, Computer Science & Engineering
Kathir College of Engineering
Wisdom Tree, Avinashi Road, Neelambur, Coimbatore-641062

Author A Author B
Aravindan Ramanathan
Ankit Singh
Dept Of Computer Science,
Tamil Nadu Agriculture University, Coimbatore. Faculty of Computing & Agriculture,
[email protected] AgriInnovate University, USA
[email protected]

ABSTRACT:

This paper introduces a paradigm shift in agriculture through deep learning

empowered leaf disease detection. Conventional methods of disease identification in crops
are time-consuming and prone to errors, impeding timely interventions. Our approach
harnesses the capabilities of deep learning to accurately detect leaf diseases, offering a robust
and efficient solution. Trained on extensive datasets, our model exhibits superior
performance compared to traditional methods, ensuring high accuracy and reliability. Its
adaptable architecture facilitates seamless integration into existing agricultural systems. This
innovation heralds a new era in disease management practices, promising to optimize crop
yield, mitigate economic losses, and fortify global food security, thus laying the foundation
for sustainable agricultural development.

Keywords: Deep learning, leafdisease detection, Agriculture, disease management,

sustainable development, global food security.

103 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

GANVIZ: TEXT-BASED IMAGE GENERATION WITH

GENERATIVE ADVERSARIAL NETWORKS
Kushagra Vaibhav#1 , Manish Kumar#2 , Abhijeet Kumar #3
#Department of Computer Science & Engineering
Sengunthar Engineering College, Perundurai, Erode

ABSTRACT:
In recent years, Generative Adversarial Networks (GANs) have emerged as a
powerful tool for generating realistic images from textual descriptions. This paper introduces
GANViz, a novel framework for text-based image generation utilizing GANs. By harnessing
the adversarial training paradigm, GANViz leverages the synergy between text embeddings
and image generation networks to produce highfidelity images from textual input.

Keywords—GAN,C-GAN,Bridging,Synthesizing,Computer Vision

104 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

FORECASTING VARIOUS CHILDBIRTH METHODS USING MACHINE

LEARNING

Yazhini S(Assistant Professor), Deepan M S, Karunagaran A, Niranjan Kumar S V

Information Technology
NandhaEngineering College-Erode

ABSTRACT:
Childbirth is a multifaceted process influenced by various factors, including maternal
health, fetal conditions, and labor dynamics. Accurate prediction of the mode of delivery,
whether cesarean section or vaginal birth, is pivotal for ensuring optimal maternal and
neonatal outcomes. We employed a combination of Support Vector Machines (SVMs) and
Random Forests (RFs), two powerful machine learning algorithms, to enhance prediction
accuracy. SVMs excel in classifying data into multiple classes, while RFs, an ensemble
learning algorithm, integrate predictions from multiple decision trees to produce more robust
forecasts. Our results indicate that the integration of SVMs and RFs significantly improves the
accuracy of predicting the mode of childbirth compared to individual algorithms. The
combined approach of SVMs and RFs effectively captures the complexity of childbirth data,
offering valuable insights for clinicians to make informed decisions. This predictive model not
only enhances clinical decision-making but also empowers expectant mothers by providing
insights into their birthing options. By advancing the capabilities of predictive analytics, this
study contributes to improving healthcare quality, reducing unnecessary interventions, and
enhancing maternal and neonatal safety.

Keywords: Childbirth Prediction, Mode of Delivery, Support Vector Machines (SVMs),

Random Forests (RFs), Machine Learning Algorithms, Maternal Health, Neonatal Outcomes,
Clinical Decision- Making, Predictive Analytics, Obstetrics, Healthcare Quality, Maternal
Safety, Neonatal Safety, Informed Decision-Making, Ensemble Learning.

105 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

DETECTION OF VIRTUAL PRIVATE NETWORKS

ENCRYPTED TRAFFIC USING MACHINE AND ENSEMBLE
LEARNING ALGORITHMS
Sujatha S1, Shivani P R2,Sakthi PriyaC3, AhilaR4
1
Assistant Professor,234UG Students
Department of Computer Science and Engineering,
School of Engineering, Avinashilingam Institute for Home Science and Higher Education for Women
Coimbatore,TamilNadu,India.

ABSTRACT:
The widespread adoption of Virtual Private Networks (VPNs) for secure communication
over public networks has highlighted the need for accurate detection of encrypted network
traffic. This study proposes a comprehensive approach utilizing machine and ensemble
learning algorithms to address this detection challenge. Leveraging the ISCX VPN dataset,
the project aims to develop a robust system capable of distinguishing between VPN and non-
VPN traffic with high accuracy. Key stages include data acquisition, preprocessing, feature
extraction, and model training using Decision Tree, Naive Bayes, Random Forest, and
Adaboost algorithms. Through rigorous experimentation, the most effective algorithmic
approach is identified. The project also offers insights into the diverse nature of encrypted
network communication and its implications for network security. The anticipated outcomes
include improved understanding and management of VPN traffic, enhancing overall network
security and performance. This research contributes to advancing network security practices
by offering practical solutions for encrypted traffic detection in real-world settings.

Index Terms— Virtual Private Network (VPN), Machine Learning, Ensemble

Learning,Traffic Analysis, ISCX VPN Dataset.

106 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

UNLOCKING THE POTENTIAL OF INDUSTRIAL

AUTOMATION
Krithicroson R,gowsidharan S T
Artificial Intelligence and Machine Learning
M kumarasamy college of engineering

ABSTRACT:
Industrial automation plays a critical role in optimizing performance within the automotive
manufacturing sector. However, existing systems encounter significant challenges, including
imprecise part classification and excessive manual intervention, leading to operational
inefficiencies and downtime. In response, System introduces a tailored automation system
designed specifically for the automotive industry. Current systems face obstacles related to
inconsistent part categorization and reliance on manual intervention, hindering productivity
and increasing error risks. This innovative system integrates advanced machine learning
methodologies, such as transfer learning, alongside real-time computer vision technology to
achieve accurate and consistent part classification seamlessly within existing production
lines. By harnessing for precise part identification and classification, our solution overcomes
these challenges, facilitating seamless integration with manufacturing lines while reducing
manual intervention. With a targeted accuracy improvement from 78% to 86%, this proposed
system embodies a paradigm shift in automotive production automation. Through reduced
manual intervention and optimized workflow methods, this system enhances efficiency and
productivity throughout automotive manufacturing operations. The discernible benefits of
this system include improved accuracy, reduced downtime, and overall enhanced efficiency,
resulting in significant cost savings and a strengthened competitive edge within the
automotive manufacturing landscape. In comparison to existing systems, this solution offers
superior precision, adaptability, and efficiency, marking a significant advancement in
automotive automation technology. This project represents a concerted effort to address
prevailing challenges within automotive production through the implementation of an
advanced automation system. By leveraging modern technology and innovative
methodologies, we aim to redefine the operational landscape of automotive manufacturing,
driving forward development and fostering sustainable growth within the industry.

107 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

ONLINE BLOOD BANK USING CHATBOT

1 S.Anukiruba, 2 R.Ragadevi, 3 A.Sajetha, 4 G.Sowbarnika, 5 C.Prakash Kumar.

1,2,3,4UG Students, Department of Computer Science Engineering,

5 Assistant Professor(CSE), 1,2,3,4,5 Angel College of engineering and Technology,Tirupur.

ABSTRACT:
This project proposes the development of an AI-driven online blood bank system
integrated with a chatbot interface to streamline the process of blood donation, matching
donors with recipients, and handling blood requests. The system aims to address the
challenges faced by traditional blood banks, such as inefficient communication, limited donor
reach, and delayed response times. The chatbot interface will serve as the primary interaction
point for users, allowing them to register as donors, search for blood donors based on specific
criteria, submit blood requests, and receive real-time assistance. By combining the power of
AI technology with an online platform, this project aims to enhance the accessibility,
efficiency, and effectiveness of blood donation and distribution processes, ultimately saving
lives and improving healthcare outcomes. The proposed AIenhanced online blood bank
system will offer several key features to improve the blood donation and distribution process.
Firstly, it will allow individuals to easily register as blood donors through the chatbot
interface, providing their relevant information such as blood type, location, and availability.
This ensures a comprehensive database of potential donors. Secondly, the chatbot will
facilitate seamless communication between donors and recipients. Donors can receive
notifications about urgent blood requests in their vicinity and respond promptly through the
chatbot. Recipients, on the other hand, can submit their blood requests via the chatbot,
specifying their blood type and quantity needed. The system will then intelligently match
donor profiles with recipient requirements, prioritizing factors such as blood type
compatibility and proximity. Furthermore, the chatbot will provide support and guidance
throughout the entire donation process, including scheduling blood donation appointments,
providing information about donation centers, and answering frequently asked questions.
This personalized and accessible approach aims to encourage more individuals to participate
in blood donation activities. Keywords: Chatbot, Blood donation, Blood request handling,
Database Management, Donor registration, Recipient notification.

108 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

FOOD CALORIES DETECTION USING GEMINI API

Aditya Baibhav#1, Md Arif Reza#2, Deepak Kumar#3
# Computer Science & Engineering dept, Erode Sengunthar Engineering College,
Perundurai
ABSTRACT:
In an era where dietary choices significantly impact health and wellness, the
accurate assessment of calorie intake is of paramount importance. "DeepCalorie" presents an
innovative food calorie detection device that leverages state-of-the-art . This device aims to
empower individuals with the ability to effortlessly estimate the calorie content of their meals
by simply capturing images. The system utilizes a diverse dataset of food images, deep neural
networks for feature extraction, and sophisticated algorithms to estimate calorie counts.
Furthermore, it offers the option to assess portion sizes, providing a more comprehensive
dietary analysis. The "DeepCalorie" device, with its user-friendly interface, represents a
promising step towards promoting healthier eating habits and fostering nutrition awareness.
This paper details the development process, challenges encountered, and the potential impact
of this technology on improving dietary choices and overall well-being. Introduction In
today's fast-paced world, where time is a precious commodity, dietary choices often take a
back seat. Amid busy schedules and numerous food options, people are increasingly looking
for convenient ways to manage their nutritional intake. The need for quick and accurate
information about the calorie content of foods has never been more critical, considering the
rising concerns about obesity and diet-related health issues. This is where technology,
particularly machine learning, can play a transformative role. The "DeepCalorie" project
represents a remarkable endeavor in the field of nutrition and technology convergence. This
project, anchored in the computer vision, aims to provide a novel solution to the age-old
problem of tracking calorie intake. DeepCalorie is more than just a technical innovation; it is
a potential game-changer in the way we approach nutrition and dietary management.

Keywords— Deepcalories , endeavor , fostering, dietary, amid, obesity

109 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

SMART TRAVEL PLANNER WEBSITE FOR TOURIST

USING JAVASCRIPT
1 R .Deepa, 2 G. Dhanalakshmi, 3 M. Vasuki, 4 A. Lakshma Vignesh, 5 S. Indhumathi
1,2,3,4 UG Students, Department of Computer Science and Engineering, 5 Assistant
professor(CSE) 1,2,3,4,5 Angel College of engineering and Technology,Tirupur.

ABSTRACT:
Tourism is gradually becoming the first biggest industry in the world. As the
foundation of tourism, tourism resources managements are the indispensably part of the
tourism system and it is of great importance to ensure the sustainable development of tourism
industry. In the era of information, effective utilization of the information technology can
enhance the second revolution of tourism. Currently tourists & traveller waste a lot of time to
plan their trips, and available all options to plan a trip right now does not actually provide a
cost effective and all substitute to their travel. This project “tour AND TOURISM control” is
used to automate all procedure of the travel and tourism, which deals with creation, booking
and affirmation and person info. The undertaking is designed HTML-Hypertext Preprocessor
as front cease and Microsoft sq. Server 2008 as backend which fits in any browsers. The
coding language used HTML and php. tour and tourism management device is used to e-book
a excursion from everywhere within the world by using a single dynamic internet site in order
to help the consumer to recognise all approximately the places and tour information in a
single website. The admin can upload programs to the internet site from a sure journey
retailers and hotels by using create a tour web page. Then the users can check in and book
every mission, they may be showed by way of the admin of their control reserving web page.
The person can see the confirmation of their my booking web page. it's far an easiest platform
for all tourists which can be without problems booked and recognise the all details. key
phrases: journey and tourism management, tour programs, tourism, package booking.

Keywords: Login, Schedule, Hotel booking, Guide booking, Flight booking, Cab booking,
Photography booking, Memories creation and send to travellers email

110 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

WINDOWS 14:UI AND UX DESIGN INNOVATION

Kishore. R , Arunachalam. M, Lokesh .M , Vijitha.S
Department of CSE
Vel’s institute of science technology and advanced studies ( vistas), Chennai
,

ABSTRACT:
The digital landscape continually evolves, and with it, the expectations of users for
seamless and enriching experiences. In response to this ever-changing landscape, the design
of user interfaces (UI) and the quality of user experiences (UX) have become pivotal for
operating systems. "Windows 14: UI and UX Design Innovation" is a project that seeks to
advance the state of UI and UX in the Windows operating system to meet the demands of
contemporary users.This project sets out to address the need for innovation in UI and UX
design by focusing on Windows 14, the latest iteration of the iconic Windows OS. The
objectives of this project are to enhance the overall user experience, provide a visually
pleasing interface, and streamline user interactions. We aim to create an operating system that
is intuitive, accessible, and adaptable to a wide range of user needs and preferences.Our work
involves comprehensive user research, including surveys, interviews, and usability testing, to
understand the needs and pain points of our target audience. By utilizing industry-leading
design tools and methodologies, we have created innovative UI elements that prioritize
clarity, simplicity, and aesthetics. Accessibility considerations are integral to our design
approach, ensuring that the Windows 14 interface is inclusive and user-friendly for all.

111 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

DESIGN AND PERFORMANCE ANALYSIS OF COMPACT

BANDPASS FILTER FOR C BAND APPLICATIONS
Mrs.Revathi.G.M.E,(Ph.D)., Ms.Sivasakthi R.(B.E)., Ms.Sudharshinika M.(B.E),
Ms.Vivekka R.(B.E)
Electronics and Communication Engineering K.Ramakrishnan college of
Technology,Trichy,TamilNadu,India

ABSTRACT:

This abstract presents the design of a compact C-band filter with emphasis on low
insertion loss and reduced noise. Leveraging advanced engineering techniques, the proposed
filter achieves optimal performance while maintaining a compact footprint. Through
meticulous design and optimization, the filter offers enhanced signal purity and minimal
signal degradation, making it ideal for Cband applications requiring high fidelity and
efficiency. The combination of low insertion loss and reduced noise ensures superior signal
integrity, while the compact size enables seamless integration into space-constrained systems.
This filter represents a significant advancement in C-band filtering technology, catering to the
demands of modern communication systems .

112 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

IoT ENABLED NON-INVASIVE BIOIMPEDANCE

SPECTROSCOPY BASED HUMAN HYDRATION MONITOR
FOR DIALYSIS PATIENTS
PRAKASH S,LAVANYA G
BIOMEDICAL ENGINEERING,SRI RAMAKRISHNA ENGINEERING COLLEGE

ABSTRACT:

Monitoring an individual’s hydration status is crucial since water plays a major role in the
regulation of the human body. Dehydration affects people of all ages, which majorly include
athletes, farmers, and ICU patients. Overhydration is seen in edema patients, patients
undergoing dialysis, and athletes who intake fluids to avoid dehydration. Heat exhaustion
occurs especially in farmers, leading to loss of fluid through Dehydration. Dehydration is a
serious problem in children and also estimated that more children will be malnourished in the
future. Malnutrition makes the child susceptible to infection and complements its effect in
contributing to child mortality which accounts for 22% of the burden of disease in India and
adversely affects the economic growth of the country with an estimated adult productivity
loss of 1.4% of gross domestic product. Hydration assessment of a person is made in a
clinical setting mostly with clinical symptoms presented by the patient. This establishes the
need for a continuous and real-time hydration status monitoring device. Thus, an IoT-enabled
Non-invasive Bioimpedance Spectroscopy based human hydration monitor was developed
for monitoring hydration states by calculating TBW, ECW, and ICW. This device helps in
providing self monitoring, early diagnosis, and better treatment which can promote good
health.

Keywords—BioImpedance Spectroscopy (BIS), Total Body Water (TBW), ExtraCellular

Water (ECW), IntraCellular Water (ICW).

113 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

CARE CONNECT APP FOR ORPHANAGE CHILDREN

C Trisha* , Mrs.Jesintha J M.sc.,
Department of Computer Science
Sathyabama Institute of Science and Technology Chennai-600119.
ABSTRACT:

Care connect is an Android app for orphanage and old-age people don't have parents. The
Orphanage Connect Android app aims to bridge the gap between orphanages, caretakers, and
potential supporters The app facilitates instant communication between orphanage staff,
volunteers, and supporters through a secure messaging system Push notifications ensure that
important updates, events, and urgent needs are promptly conveyed to all relevant
stakeholders .Orphanages can list their specific needs, such as food, clothing, educational
materials, and financial support The app supports educational programs by providing a
platform for sharing learning resources, organizing virtual classes, and connecting students
with mentors connect and interact with them. An integrated calendar system for scheduling
and managing medical appointments, check-ups, and follow-up visits. A feature to track and
manage medications, including dosage schedules and reminders for caregivers to administer
medications. Integration with a donation portal to facilitate contributions for medical
supplies, medications, and healthcare services. Customization to account for cultural and
regional differences in healthcare practices, ensuring that the app is culturally sensitive and
relevant. This app will connect the gap between who don’t have parents and caretakers and
we can help them with giving education and extracurricular activities etc.

Keywords— Care connect,Orphanage Connect Android app, caretakers, and potential

supporters ,donation,schedules.

114 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

TRANSPARENT ELECTRONIC VOTING SYSTEM USING

BLOCKCHAIN,
MRS.N.DIVYAA B.E.,M.E.,( ASSISTANT PROFESSOR), SARMILA.K, SRIHARIHARAN.G,
SURNDHAR.G, VASANTHA KUMAR R
DEPARTMENT OF ECE
VELALAR COLLEGE OF ENGINEERING AND TECHNOLOGY, ERODE

ABSTRACT:
People are seeing their basic rights violated underscoring the essential nature of
democratic processes around the world, as distrust of traditional electoral systems grows.
Difficulties with alternative digital voting systems, especially their lack of trust the
transparency of the democratic process. Compounding the problem is the fact that both old
and modern digital methods are prone to abuse, underscoring the urgent need for solutions to
ensure the integrity and integrity of electoral processes. The emergence of blockchain
technology is a viable solution to these problems and supports democratic values. Using
blockchain technology, we can create a transparent and fair voting system that build trust
between voters and election officials. As it is based on blockchain technology, our proposed
platform creates a strong foundation for secure and efficient digital voting, eliminating
physical inefficiency evaluation, which shows that system can be adopted on a large scale. a
viable solution to ensure democratic rights in today's societies.

KEYWORDS - 1. Fundamental Rights 2. Transparency of Democratic Process 4. Blockchain

Technology 4. Digital voting systems 5. Democratic rights.

115 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

ANOMALY DETECTION BASED ON MACHINE LEARNING

IN WIRELESS NETWORKS
Pavithra S S
Post Graduate Student
Department of Electronics And Communication Engineering
BannariAmman Instituteof Technology,Sathyamangalam.
[email protected]

ABSTRACT:

The rapid expansion of PC networks has led to challenges in maintaining network

availability, integrity, and confidentiality. As a result, network administrators are increasingly
turning to anomaly detection systems to monitor network traffic for unauthorized and
malicious activities. Anomalies represent breaches of security policy with malicious intent.
Therefore, anomaly detection systems monitor traffic flowing through network computers to
identify malicious activities and known threats, alerting administrators when such threats are
detected. Anomaly detection, which identifies malicious activity as any behavior that deviates
from normal patterns. This paper provides an overview of various efforts aimed at
constructing efficient CNN network detection systems using multiple machine learning (ML)
classifiers. The effectiveness of these systems is evaluated using KDD Cup’99 datasets. The
paper discusses and compares the results obtained by various studies, offering insights and
guidance for future research in this area.

Keywords— Anomalies Detection, Machine Learning, security, Convolutional Neural

Network (CNN), Misuse detection, Multiple Classifiers.

116 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

DEEP LEARNING-DRIVEN BEARD RECOGNITION

SYSTEM WITH AUTOMATED ALERTS
M. Pavithra B.E. M.E(Assistant. Professor), Amuthapriyaa G L, Gopika G K, Aarthi R,
Devadharshini S
Computer Science and Engineering
K.RAMAKRISHNAN COLLEGE OF TECHNOLOGY,TRICHY, INDIA

ABSTRACT:

Facial recognition systems are crucial for security and surveillance applications.
However, their performance can be significantly hampered by variations in facial appearance,
particularly the presence of beards. This paper proposes a novel deep learning-driven beard
detection system with automated alerts that addresses this challenge. The system employs a
Convolutional Neural Network (CNN) architecture to effectively classify faces with and
without beards, even under diverse lighting conditions and beard styles. Additionally, it
integrates automated alert mechanisms that notify security personnel of individuals with
beards who might be flagged for security reasons or exhibiting suspicious behavior. This
approach offers several advantages over traditional facial recognition, bolstering overall
security posture.

Keywords—Beard Detection, Deep Learning, Convolutional Neural Network (CNN), Facial

Recognition, Security, Surveillance, Automated Alerts.

117 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

SMART PLANT MONITORING SYSTEM USING IOT

Authors: Yogeshwaran VB, Vikram N Co Author: Nirmal A
Department of ECE,
K Ramakrishnan College of Technology, Samayapuram, Trichy.

ABSTRACT:

In this smart and automated world, everything has become technically smart and
intelligent due to development of advanced devices. In this article we have make our plants
also smart with the use of some technical components so that they can maintain themselves
without any human help. This system is designed to maintain the moisture content and
temperature in the plants. The system will also be able to alert the user if any of these factors
go beyond a certain range and activate the water pump to irrigate the plant. Additionally, the
system will be able to detect motion in the environment where the plant is growing and alert
the user. The system is built using an ESP8266 Node MCU board, a DHT11 temperature, a
soil moisture sensor, a 18650 battery, a relay module, and a water pump.

118 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

EFFICIENT ERROR DETECTION SCHEMES FOR ECSM

WINDOW METHOD
Authors: Sakthi vishwa J L, Thevanand B Co-Author: Rakesh S
Department of ECE,
K Ramakrishnan College of Technology, Samayapuram, Trichy
ABSTRACT:

Elliptic curve scalar multiplication (ECSM) is a critical component in elliptic curve

cryptography (ECC), a cornerstone of pre-quantum public key cryptography. However,
hardware implementations of cryptographic systems using ECSM are susceptible to both
permanent and transient errors. Ensuring the correctness of computations performed on
hardware or software is crucial for cryptographic systems to detect such errors. This article
presents novel fault detection schemes tailored for window method scalar multiplication, a
domain that has not been extensively explored previously. Our approach introduces refined
algorithms and implementations capable of mitigating both permanent and transient errors
effectively. We evaluate the efficacy of our approach through extensive simulations under a
fault model, ensuring the results reflect real-world scenarios. The proposed schemes achieve
comprehensive error coverage.

119 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

SECURITY AND PRIVACY FOR HEALTHCARE

Mrs.S.Thangamani M.Tech (IT), PhD, Deepak . N, Jeevaprasanth.V, Mohammed sarjun.S
Information Technology
Nandha Engineering College
,
ABSTRACT:
Block chain technology has the potential to completely transform healthcare
management systems by enhancing data security, interoperability, and integrity because to its
decentralized and unchangeable structure. Sensitive patient information is more difficult for
bad actors to breach thanks to the SHA-256 algorithm, which improves data security.
Interoperability between diverse healthcare systems is facilitated by smart contracts and
standardized data formats. However, issues with user adoption, scalability, and regulatory
compliance also face block chain implementation in the healthcare industry. This study offers
important insights into the changing environment of healthcare data management by
highlighting the possible advantages and challenges of integrating block chain technology in
healthcare management systems.

Keywords: patient monitoring, health, security, EHRs, Block Chain

120 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

PLANT DISEASE CLASSIFICATION FROM LEAF IMAGES

USING SVM
Mr. M Raja, Ms. RA Akshayaprada, Ms. P. Dharani, Ms. V. Miruthula, Ms. S. Sandhiya
Electronics and Communication Engineering
K Ramakrishnan College of Technology Samayapuram, Trichy.

ABSTRACT:

In agricultural ecosystems, the timely and accurate detection of plant diseases is

crucial for ensuring food security and maximizing crop yield. This study proposes a robust
approach for plant disease classification from leaf images, focusing on five common diseases:
Anthracnose, Bacterial blight, Gray mold, Citrus canker, and Powdery mildew. The proposed
methodology leverages Support Vector Machine (SVM) algorithm, a powerful supervised
learning technique known for its effectiveness in classification tasks. Initially, leaf images
exhibiting symptoms of the aforementioned diseases are collected and preprocessed to extract
relevant features. These features capture distinct characteristics such as patterns,
discoloration, and texture variations, which are crucial for disease differentiation.
Subsequently, the SVM classifier is trained using the extracted features to build a predictive
model capable of accurately identifying the presence of specific diseases in unseen leaf
images The proposed methodology holds significant promise for practical implementation in
automated agricultural systems, aiding farmers in timely disease diagnosis and management,
thus contributing to enhanced crop productivity and sustainable agriculture.

Keywords: plant diseases classification, SVM, microwave measurements

121 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

DESIGN AND IMPLEMENTATION OF SMART HOME

AUTOMATION SYSTEM USING RASPBERRY PI
P. Nalina, R. Chitra (Assistant Professor(SS /ECE))
Department of ECE, School of Engineering
Avinashilingam Institution for Home science and Higher Education for Women, Coimbatore.

ABSTRACT:

A large number of appliances are now being connected to each other on a daily
basis as a result of the Internet of Things (IoT) becoming more and more prevalent. The
increase in connectivity offers the possibility of improving a person's quality of life and
making their lives easier and more convenient. IoT has huge impacts on different aspects of
people's lives, including the home, which can create a safe environment for their family and
make their day-to-day activities more convenient. The main objective of this system is to
provide a centralized platform for automated gas leak prevention, lighting control and water
tank filling using Raspberry pi Pico. This concept increases the security, energy efficiency
and comfort of the house by connecting multiple sensors and actuators.

Keywords—Internet of Things (IoT), Smart Home, Home Automation, Raspberry Pi Pico.

122 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

AUTOMATED WATER SURFACE CLEANING SYSTEM

USING ESP8266
Ms.Anitha S(Assistant professor), Swetha S, Vignesh SV, Anjali parmar, Kaviya V
Electronics & communication Engineering
KGiSL Institute of technology,Coimbatore

ABSTRACT:

This paper presents the design and implementation of an automated water surface
cleaning system. The system integrates an ESP8266 microcontroller for data processing and
communication, an ultrasonic sensor for distance measurement, motors for movement
control, and a solar panel for sustainable power supply. The proposed system aims to address
the challenges of maintaining clean water surfaces efficiently and autonomously. The
ESP8266 microcontroller facilitates wireless communication and control, enabling real-time
monitoring and management of the cleaning process. The ultrasonic sensor accurately detects
the water surface level, allowing the system to adjust cleaning operations based on the
detected data. The motors are utilized for maneuvering the cleaning mechanism across the
water surface, ensuring comprehensive coverage and effective cleaning. Additionally, the
integration of a solar panel provides the system with renewable energy.

Keywords— Automated Water Surface Cleaning, ESP8266, Ultrasonic Sensor, Motors, Solar
Panel.

123 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

MANUFACTURING OF AMMONIUM SULPHATE BY

DIRECT NEUTRALIZATION

Ashwin Kumar .A,A.Vijayaragavan,Karthik Ganesh.R

BTech Chemical Engineering ,Kalasalingam Academy of Research and Education

ABSTRACT:

Ammonium sulphate is a common nitrogen fertiliser that is made by a number of

processes, one of which being direct neutralisation. An outline of the main procedures,
benefits, and environmental effects of the direct neutralisation method of producing
ammonium sulphate are given in this abstract. Ammonium sulphate [(NH₄)₂SO₄] is created
when ammonia gas (NH₃) reacts with sulfuric acid (H₂SO₄) at the start of the process. Since
the neutralisation reaction is exothermic, precise temperature and reaction condition control
are necessary to provide the best possible conversion and product purity. Solid ammonium
sulphate is produced by concentration and crystallisation of the resultant solution after
neutralisation.
High product purity, effective use of raw resources, and relatively simple process control are
only a few benefits of direct neutralisation. It also makes it possible to use ammonia from a
variety of sources, such as industrial operations and agricultural waste products, which
improves resource efficiency and lowers waste. The production of ammonium sulphate
through direct neutralisation has certain environmental issues, chiefly associated with the
release of ammonia and sulfuric acid into the atmosphere. While sulphuric acid emissions
may cause acidification of soil and water bodies if improperly handled, ammonia emissions
can contribute to air pollution and have effects on ecosystems and human health. Thus, to
lessen the process's negative environmental effects, efficient emission control methods like
scrubbers and pollution monitoring systems are essential.In summary, the production of
ammonium sulphate through direct neutralisation presents a flexible and effective method for
yielding a useful nitrogen fertiliser. To guarantee sustainable manufacturing methods and
reduce detrimental effects on the environment and public health, however, great attention to
environmental issues is required.

124 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

DETECTION OF HEAVY METALS USING ELECTROCHEMICAL

METHODS
Sneha P,Deepa B Prabhu
Biomedical Engineering,Sri Ramakrishna Engineering College

ABSTRACT:

The impact of heavy metals on humans has been significant, with lead particularly
posing grave health risks. Exposure to lead can lead to various health issues, including
neurological disorders, and developmental delays and also increases the risk of Cancer. Early
detection of heavy metal toxicity is crucial for preventing underlying diseases. One promising
approach for heavy metal detection is through non-invasive methods, such as urine analysis.
Urine serves as a reliable indicator of heavy metal exposure as these toxic substances are
excreted through the kidneys and accumulate in the urine. This method provides greater
accuracy as it directly reflects the body's burden of heavy metals. Heavy metals tend to
accumulate in various body tissues over time, creating detrimental effects. The main
objective of this project is the development of a novel electrochemical device utilizing boron-
doped diamond to instantaneously detect and quantify the presence of lead ions in urine,
using artificial urine for testing. Employing certain voltammetry techniques at different parts
per million (ppm) levels in liquid samples ensures accuracy in detection. Additionally, the
development of miniaturized potentiostat enables significant accessibility for portable
detection, making it easier to screen for heavy metal exposure in various settings.

125 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

HUMAN EYE BALL GAZING AND FACE-BASED ULTRA

DIGITAL SECURITY SYSTEM USING AI

D.R. Divya1, Abarnaa T2, Priya Dharshini T3, Susmitha S4

1
Assistant Professor, 2,3,4 UG Students
Department of Computer Science and Engineering
School of Engineering, Avinashilingam Institute for Home Science and Higher Education for
Women, Coimbatore, Tamil Nadu, India

ABSTRACT :

The "Eye Ball Gazing and Face-Based Ultra Digital Security System" pioneers a
formidable security approach by seamlessly integrating eye ball movement tracking, facial
recognition, and multi-factor authentication. During enrollment, it captures and securely
stores individuals' distinct eyeball-gazing patterns and facial features, creating a robust
database of unique identifiers. When access is sought, the system compares real-time eye-ball
gazing data to the stored patterns, simultaneously verifying facial characteristics, significantly
reducing the likelihood of unauthorized access. This dual authentication process provides
enhanced security, as it requires two separate biometric factors to be validated before access
is granted, offering advanced levels of protection against unauthorized access while ensuring
auser-friendlyexperience.

Keywords: Image Processing, Machine Learning, Eye-tracking technology, Gaze-based

control.

126 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

DETECTING LUNG CANCER IN CT SCANS USING CNN: A

DEEP LEARNING APPROACH
Mrs.Kiruthika D(Assistant Professor), Mukesh D, Sivasakthi V, Yokeshwaran J
Information Technology,Nandha Engineering College,Erode
ABSTRACT :
Lung cancer stands as one of the deadliest diseases worldwide, highlighting the urgent
need for early detection and precise treatment strategies to combat its high mortality rates.
Utilizing computerized tomography (CT) scans, especially in conjunction with deep learning
models, has emerged as a pivotal approach for detecting lung cancer at its nascent stages.
This study proposes a framework based on Convolutional Neural Networks (CNNs) for early
lung cancer detection using CT scan images. Furthermore, we undertake a comparative
assessment with leading models such as Inception V3, Xception, and ResNet-50. Our
analysis encompasses key metrics including accuracy, Area under Curve (AUC), recall, and
loss. Our findings reveal CNN's superiority, indicating its potential over traditional methods.
Notably, CNN achieves an accuracy rate of 92%, an AUC of 98.21%, a recall of 91.72%, and
a loss of 0.328, underscoring its promise in enhancing lung cancer diagnostic capabilities.

Keywords: Lung cancer, deep learning, convolutional neural network, CT scan images, early
detection, comparative analysis, Inception V3, Xception, ResNet-50, accuracy, AUC, recall,
loss.

127 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

AN EXPERIMENTAL STUDY ON MECHANICALPROPERTIES OF

BORASSUS FLABELLIFER

Dr.T.VarunKumar1,M.Manoj2,M.Nishanth3,P.Prasanth4,N.Santhosh5
1
-Professor,DepartmentMechanical Engineering,P.ACollegeofEngineeringandTechnology.
2
-DepartmentofMechanicalEngineering,P.A.CollegeofEngineeringand Technology.

ABSTRACT:

The palm tree species Borassus flabellifer, also referred to as the Palmyra palm or
toddy palm, is indigenous to Southeast Asia and the Indian subcontinent. This towering,
lonely tree is well known for its many applications in a wide range of fields and cultural
contexts. Large clusters of edible fruit with a fibrous, sweet pulp are produced by the tree
and are used in cooking. Fur- thermore, the sap of the Palmyra palm, sometimes referred to
as toddy, is collected and consumed as a cool beverage or fermented to create alcoholic
drinks. The leaves, husk, and wood of the tree are just a few of the parts that are useful in
traditional crafts, building, and other activities. Nowadays, Borassus flabellifer fibers are
used as a reinforcement in epoxy composites for structural purposes is investigated in this
work. The toddy palm, or Borassus flabellifer, is a naturally occurring fiber with desirable
mechan- ical qualities that make it a desirable option for composite materials. Fibers from
Borassus flabellifer were collected, processed, and then mixed into an epoxy matrix to
create composite specimens. The mechanical qualities like tensile, impact, and load, are
examined in American society for testing and materials (ASTM).

Keywords - Borassus flabellifer, Alkaline treatment, ly556 resin, hy551 hardener, CTF
fibers, Epoxy resin

128 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

DETECTION OF FRADULENT ACTIONS IN CARD

TRANSACTIONS
Uma.J1 (Assistant Professor), Gopika.M2 Hemapriya.P3 Abirami.S.P4
Computer Science and Engineering Department,Jai shriram Engineering College,Tirupur.

ABSTRACT:
The unauthorised and unwelcome access to a debit or credit card by a person who isn't
the account owner is fraudulent in credit card transactions. The identification of unauthorized
credit card transactions is currently the most common problem. The development of e-
commerce websites and internet transactions is to blame for this. Credit card fraud typically
happens when a card is taken and utilized for any questionable activity, or if a con artist uses
the cardholder's details for his own advantage. These days Employing phishing technologies
to commit internet banking fraud entails sending or stealing funds and account information
without the banker's consent. In the current era of mobile payments, credit card fraud
detection is a crucial study. Due to the frequent changes in both user payment behaviour and
criminal fraud conduct, it is extremely challenging to increase a fraud detection model's
effectiveness and preserve its stability. The technique for detecting credit card theft was
created to recognize fraudulent actions. At the same time, we work to prevent genuine
transactions from being confused with fraudulent ones. This paper focuses on obtaining a
deep learning algorithm oflegitimate and fraudulent transactions from a loss function
perspective.

KEYWORDS:

Artificial Intelligence

129 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

LUNG TUMOR SEGMENTATION AND STAGE DETECTION

FROM CT IMAGE
Mrs.Mohan Kumar P,M.E,Ms. Joshni Rentino J X, Ms.Yazhini M, Ms.Lourdus Cerina A
Biomedical Department, PSNA College Of Engineering and Technology Dindigul.

ABSTRACT:

The identification of lung cancer at benign stage is a formidable challenge in medical

research is addressed through the machine learning and image processing methodologies by
increasing the advanced computational techniques rooted in segmentation algorithms and
machine learning models such as those implemented using matlab ct scans are analyzed to
accurately identify and classify lung tumors as benign or malignant this approach not only
enhances diagnostic precision but also holds the promise of improving patient prognosis and
reducing mortality rates by enabling timely intervention and treatment the synergy between
image processing and machine learning presents a transformative opportunity to mitigate the
impact of lung cancer through early identification and intervention ultimately advancing
public health outcomes.

Keywords: Lung cancer, MATLAB,CT images, Segmentation, Classification.

130 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

EMPOWERING TAMIL NADU: CENTRALIZED GRIEVANCE

REDRESS AND MONITORING SYSTEM
Gayathri R(Associate Professor), R.Pijo, V.Prabhu, S.Mathan, V.Mariselvam
Department of Computer Science and Engineering
PSNA College of Engineering and Technology
Dindigul, Tamil Nadu

ABSTRACT:
Recent experiences show that deficiencies in a complaint management system
could have massive costs in time, money and damage to the individual careers. The purpose
of this project is to maintain the discipline and cleanliness inside the city using a website. It is
hard for the management to look into every issue that arises inside the city. If city have issues
like electricity, water, drainage, road management, garbage and face struggle in maintaining.
This platforms helps the users to deliver the complaints and to make it rectify.Empowering
TN utilizes Automation, React JS, Application Programming Interface(API) and Cloud
technologies to engage users to complaint the fault, track live location of complaints for
admins and track status of complaint for user. We added additional feature of multilingual
support, department wise admin and automated mail notification.The public users post
complaints with geo-tagged media along with details. In our website we maintain a separate
login for administrators of different departments. When the user posts a complaint in a
specific department the complaint is only visible to the admin of that particular department.
This followed for every city. After the issue has been posted on the website the intermediate
person will check the complaint. Thecomplaint also get automatically notified to the
particular department’s mail. The user can view the live status of a complaint like “Unseen,
Seen, Ongoing, Rectified” on the user’s dashboard. For admin to verify the complaint we
added “Rectified complaints, Unrectified complaints” on admin dashboard. For user safety
we analyzed and added delete complaints only for user and not for admin.
Overall, the "Empowering Tamil Nadu: Centralized Grievance Redress and
Monitoring System" serves as a powerful tool for promoting citizen participation, enhancing
administrative efficiency, and fostering transparency and accountability in governance. By
empowering citizens to make complaints about the issues and avail themselves of
government benefits, the Empowering TN contributes to the overall well-being and social-
economic development of society.

131 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

MODEL PREDICTIVE CURRENT CONTROL OF REDUCED

SWITCH MULTILEVEL INVERTER
Vibinakash.S , Bhuvaneshwari.K , Sivabharathi.KS
Department of electrical and electronics engineering,
Dr.Mahalingam college of engineering and technology, Tamilnadu, India.

ABSTRACT:
This project proposes a model predictive current control (MPCC) strategy for
a reduced switch multilevel inverter (RSMI) topology. The RSMI offers advantages such
as reduced component count and improved efficiency compared to traditional multilevel
inverters. The Model Predictive Current Control (MPCC) algorithm utilizes a finite
control set approach to predict future current errors and selects the optimal switching
states to minimize these errors while considering constraints such as voltage limits and
switching frequency. Simulation results demonstrate the effectiveness of the proposed
MPCC strategy in achieving accurate current tracking and maintaining voltage balance
across the inverter's output phases. Experimental validation on a hardware prototype
further verifies the feasibility and performance of the proposed control technique,
highlighting its potential for applications in renewable energy systems, motor drives, and
grid-connected power converters.

132 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

PCB DESIGN OF H-BRIDGE USING KICAD

Mr.T.Anandkumar, Nivash S, Baby Shalini K, Ranjith S
ELECTRICAL AND ELECTRONICS ENGINEERING,
DR.MAHALINGAM COLLEGE OF ENGINEERING AND TECHNOLOGY

ABSTRACT:
The main goal of this project is to utilize KiCad to design a printed circuit board
(PCB) for an H-bridge motor driver. Because the H-bridge architecture can drive motors in
both directions, it is frequently used in motor control applications. The layout of high-power
traces, thermal control, component location, and signal integrity are all taken into account in
PCB design. PCB layout, component footprint creation, and schematic capture are all done
with the open-source EDA application KiCad. The abstract will examine the main design
obstacles, the techniques used, and the finished PCB layout, emphasizing KiCad's
productivity and efficacy throughout the development process.

133 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

SUPPLY CHAIN MANAGEMENT ON BLOCKCHAIN

1Pradeep P,2A.Banushri,
Department ofComputer ScienceEngineering
VelsInstitute of Science,Technology&Advanced Studies Chennai,India

ABSTRACT:

Traditional supply chains face challenges like traceability, counterfeiting, and lack of
trust. Thispaper explores how blockchain technology, with its core features of immutability,
distributed ledgers, andcryptography, can address these issues. By providing a tamper-proof
record of a product's journey, blockchainenhances traceability and empowers stakeholders to
verify authenticity, origin, and ethical sourcing. Smart contracts, self-executing agreements
built on blockchain, further improve efficiency by automating tasks andstreamlining
transactions. Additionally, blockchain fosters collaboration and trust by creating a shared
platformfor data exchange, eliminating intermediaries, and providing a holistic view of the
supply chain. Whilechallenges such as scalability, regulation, and integration exist,
blockchain holds immense potential to transformsupply chain management, paving the way
for a more secure, transparent, and efficient future. Keywords: Blockchain, Traceability,
Immutability, Distributed Ledgers, Smart Contracts, Supply ChainManagement, Authenticity,
Cryptography, Ethical Sourcing, Efficiency, Data Exchange, Transparency

134 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

ACCIDENT DETECTION AND ALERT SYSTEM

S.Keerthana, R.Ratchaga, S.Akilesh

Department of Electrical and Electronics Engineering,

Dr.Mahalingam College of Engineeerimg and Technology, Pollachi.

ABSTRACT:
Automatic information to the ambulance and the ambulance will reach the spot fastly
can save human life and reduce number of people dying every day. The proposed system is to
reduce the death rate of human every day due to accidents by updating through the
ambulance and so can save human life. To ensure the safety of the passenger and help the
passenger if the accident occurs is the key role plays in our project. The automobile circuit
has a vibration sensor, GSM, GPS module. Vibration sensor crash or hit automatically detect
an accident location by GPS module and alert the rescue team or control station and upload
the location in IOT cloud page. The project concludes when a vehicle meets with an accident,
the vibration sensor will analyze the signal and this signal to node mcu send the location sms
by GSM module to the ambulance and rescue team.

135 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

AREA EFFICIENT MODIFIED COMPRESSOR BLOCKS

FOR IMAGE MULTIPLICATION
Dhanuversha A,Helan Sneha A, Kowsalya S
Department of Electronics and Communication Engineering
K Ramakrishnan college of Technology

ABSTRACT :

This abstract discusses the utilization of approximate computing in the design of

energy-efficient nanoscale digital circuits. By compromising absolute accuracy, circuit
designers aim to optimize parameters such as transistor count, power consumption, delay, and
area. Multiplication, known for its resource-intensive nature, can benefit from efficient
approximate multipliers that significantly reduce energy consumption and hardware
overhead. Two main methodologies are explored for designing such multipliers: integrating
approximate adders and compressors into conventional multiplier structures or introducing
structural modifications to achieve approximation. Error compensation modules can also be
employed to mitigate errors in approximate multipliers. This approach is particularly relevant
in error-tolerant applications like convolutional neural networks used in speech and image
recognition, where a certain margin of error is acceptable as long as the results are still
discernible to humans. The adoption of approximate computing presents a promising
opportunity to achieve energy efficiency without compromising the significant functionality
of nanoscale digital circuits. This abstract introduces a novel approximate multiplier design
for error-tolerant applications that combines error compensation modules and
truncationtechniques. The proposed design achieves ultra-efficient hardware with reduced
transistor count and energy consumption, while maintaining sufficient accuracy for neural
networks and image processing. The key features of hardware are efficiency, scalability, and
accuracy in error-resilient applications. Its performance is demonstrated through evaluations
using reliable benchmarks, offering promising solutions for future digital circuits with
reduced hardware overhead, delay, and power consumption.

136 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

AUTOMATIC TEMPERATURE CONTROL SYSTEM IN

POULTRY FARM
K.Karthik , R.Madhu shree , R.Gowtham
Department of electrical and electronics engineering,
Dr.Mahalingam college of engineering and technology,Tamilnadu,India.

ABSTRACT:
Poultry farming requires meticulous temperature regulation for optimal bird health
and production efficiency. This abstract presents a novel approach to temperature control in
poultry farming employing the DHT11 sensor. The DHT11 sensor, renowned for its accuracy
and affordability, is integrated into a comprehensive temperature control system. Through
real-time data collection, the DHT11 sensor continuously monitors ambient temperature
within the poultry house. Upon deviation from the desired temperature range, the system
triggers automated adjustments, such as activating cooling or heating mechanisms. This
proactive approach ensures that the temperature remains within the optimal range, mitigating
stress on the birds and enhancing overall productivity. Furthermore, the system offers remote
monitoring and control capabilities, empowering farmers to manage temperature fluctuations
efficiently and effectively. By leveraging the DHT11 sensor's capabilities, this temperature
control system represents a significant advancement in poultry farming practices, fostering
improved animal welfare and economic sustainability.

137 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

LEVERAGING SENTIMENT ANALYSIS FOR E-COMMERCE

INSIGHTS
Mr.T.Kanagasabapathy AP(SS), Gowsika V, Vivitha Shree B, Suganthan Balaji M
Department of Computer Science and Engineering
Dr.Mahalingam College of Engineering and Technology Pollachi.

ABSTRACT:
In the era of burgeoning e-commerce platforms, the analysis of customer sentiment
through reviews has become pivotal for businesses aiming to understand market dynamics
and enhance customer satisfaction. However, the sheer volume of feedback necessitates
automated techniques for efficient analysis. This paper presents a comprehensive approach to
leverage sentiment analysis for e-commerce insights. We introduce a web application tailored
to scrape product reviews from Flipkart, a prominent e-commerce hub, and subsequently
analyze sentiment to extract actionable insights. Our methodology integrates web scraping
techniques with advanced natural language processing (NLP) algorithms, facilitating efficient
sentiment analysis. The implemented system encompasses both frontend and backend
components, offering a user-friendly interface for inputting product URLs and presenting
sentiment analysis results. Through this endeavor, we aim to equip businesses with a
powerful tool to derive valuable insights from customer feedback, thereby fostering informed
decisionmaking in the realm of e-commerce.

Index Terms—Sentiment Analysis, E-commerce, Customer Reviews, Web Scraping, Natural

Language Processing (NLP), Machine Learning, Flipkart, Web Application, Data Analytics,
Customer Feedback.

138 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

STANDALONE NON-INVASIVE GLUCOSE AND

CHOLESTEROL MONITOR

L.DHIVYALAKSHMI,SHREYA.V,SWETHA.M,VARSHINI.S
Biomedical Engineering,Sri Ramakrishna Engineering College Coimbatore.

ABSTRACT:
Diabetes and cardiovascular diseases are significant health concerns globally,
contributing to substantial morbidity and mortality rates. Traditional methods for monitoring
these conditions rely on invasive procedures, leading to patient discomfort and suboptimal
disease management. Moreover, the disposal of biomedical waste from these procedures
raises environmental sustainability concerns. To address these challenges, this paper presents
a novel solution: a standalone device for non-invasive detection of both glucose and
cholesterol levels. Utilizing cutting-edge optical sensor technology, this device integrates
advanced sensors, microcontrollers, and user-friendly interfaces. By leveraging principles of
light absorption and transmission through tissues, the device provides real-time readings of
both glucose and cholesterol levels without the need for invasive blood sampling. This
breakthrough technology not only enhances patient comfort but also improves convenience
and promotes environmental sustainability in disease management. This innovative device
offers a comprehensive solution for monitoring both glucose and cholesterol levels,
empowering individuals to manage their health more effectively while reducing the
environmental impact associated with traditional invasive procedures.

Keywords:

Non-invasive detection, Optical sensor technology, User-friendly

139 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

BRAIN TUMOR DETECTION USING ML ALGORITHMS

MRS.PAVITHRA JAGADEESAN(ASSISTANT PROFESSOR), NARESH KUMAR S

COMPUTER SCIENCE AND ENGINEERING ,VELS UNIVERSITY

ABSTRACT:
This survey explores the transformative role of machine learning (ML) in brain tumor
detection from medical imaging data. ML algorithms such as support vector machines
(SVMs), random forest (RF) , and ensemble methods are systematically reviewed for
their applications and performances in tumor detection tasks. Additionally, preprocessing
techniques (e.g., denoising, normalization) and feature extraction methods (ranging from
traditional handcrafted features to deep learning-based representations) are discussed for
enhancing image quality and capturing discriminative features. Addressing challenges
like model interpretability, class imbalance, and dataset scarcity, this survey emphasizes
robust evaluation metrics and data augmentation strategies. Future directions include
integrating multimodal data for comprehensive tumor characterization, developing
explainable AI techniques, and exploring federated learning for collaborative model
training with data privacy. Ultimately, this survey underscores ML's potential to
revolutionize brain tumor diagnosis and treatment planning, guiding further
advancements in the intersection of medical imaging and machine learning.

140 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

MALICIOUS MOBILE APPLICATION DETECTION

Mr. S. Senthil Prabhu AP(SS)/(CSE), Swetha N B, Sudharsana A, Sharmila S
Department of Computer Science and Engineering
Dr. Mahalingam College Of Engineering and Technology, Pollachi, Tamil Nadu, India

ABSTRACT:
This project introduces a Malicious Mobile Application Detection System helps to
identify and classify harmful mobile apps. This project strives to achieve an impactful results
by developing a strong threat detection system capable of identifying and categorizing
potentially harmful mobile applications. Through careful preprocessing, advanced feature
engineering, and model optimization, the system achieves high performance. The
consideration of real-world use, including potential integration into mobile platforms with a
user-friendly interface, enhances accessibility and user experience. Ethical considerations
guide responsible practices in handling potentially harmful applications, and comprehensive
documentation supports knowledge dissemination. Overall, this initiative significantly
contributes to fortifying mobile ecosystems against the increasing threat of malicious
applications, prioritizing the global need to protect user privacy and security.

Index Terms—Malicious Mobile Application Detection System, threat detection, harmful

applications, preprocessing, model optimization, real-world use, integration, user-friendly
interface, ethical considerations, comprehensive documentation, fortifying ecosystems, user
privacy, security

141 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

IOT IN GESTURE - BASED CONTROL FOR SMART HOMES

REDEFINING SMART HOME INTERACTION
Ms. S. Devilavanya (AP/CSE), Dharun R, Viyaas K G E, Sangeetha K
Department of Computer Science and Engineering
Dr. Mahalingam College of Engineering and Technology Pollachi, Tamil Nadu.

ABSTRACT:
Integrating gesture recognition with the ESP32- CAM module enables intuitive, hands-free
smart home control. Leveraging the module's video processing, it interprets live feeds
through machine learning. This system recognizes specific hand movements for seamless
interaction with devices. It prioritizes user experience by offering a gesture-based interface
for home control. Mapping gestures to actions like lighting adjustments enhances
convenience. The ESP32-CAM serves as a responsive hub for real-time automation. It
reduces reliance on traditional interfaces, adding convenience to management. Security
measures ensure the integrity and privacy of the smart home ecosystem.

Index Terms: Gesture recognition , ESP32-CAM module , Smart home control , Hands-free
interaction , Machine learning ,Real-time automation , User experience , Privacy and
security.

142 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

IOT IN RETAIL FOR SMART SHELF MANAGEMENT

Mr. P. Boopathi Rajan, AP(SS), Navya. K, Anushruthi. K, Shivya. K
Department of Computer Science Engineering
Dr. Mahalingam College of Engineering and Technology Pollachi

ABSTRACT:

In today's fast-paced supermarket environment, managing shopping during peak

hours can be challenging for both shoppers and retailers. To alleviate congestion and enhance
the shopping experience, an innovative solution is proposed: an automatic trolley equipped
with RFID tags, a GSM module, and an LCD display. Shoppers can swiftly scan items using
the RFID tags, triggering instant processing through the trolley's embedded system. Real-time
updates and exclusive offer messages are displayed on the LCD screen as shoppers navigate
the store, providing them with valuable information and enhancing their overall shopping
experience. This system promises to streamline the shopping process, reduce wait times, and
engage shoppers with personalized offers, thereby improving efficiency and customer
satisfaction in supermarkets. Index Terms: Automatic trolley, RFID tags, Real-time updates,
Personalized offers, reduce wait times, efficiency.

143 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

ACTIVE PROSTHETIC ANKLE AND ADAPTIVE

EQUIPMENT FOR BIKE RIDING AND LOCOMOTION IN
LOWER LIMB AMPUTEES
Gayathri C, Dhanapal R, Dharanya R, Brindhadevi S, Sree Durkkha P A
Dept. of Biomedical Engineering, Vivekanandha College of Engineering for Women
Namakkal.
ABSTRACT:
Independent locomotion is vital for lower-limb amputees, as it not only improves
physical health but also enhances psychological well-being and social integration, enabling
them to lead fulfilling and independent lives. Therefore, developing effective prosthetic
solutions that enable independent mobility is essential for improving the overall quality of
life for individuals with lower limb amputations. Many lower-limb amputees face the
challenge of being unable to ride bikes due to the limitations of existing prosthetic systems,
which are often unaffordable and not designed for biking by middle-class people. By
addressing the current lack of affordable and suitable prosthetic solutions, this project aims to
empower lower-limb amputees to engage in biking activities, thereby enhancing their
mobility and quality of life. The objective is to enhance the functionality of the traditional
Jaipur foot to make it suitable for bike riding, thereby empowering lowerlimb amputees to
engage in this mode of transportation. The modified Jaipur foot will be engineered to provide
the stability, flexibility, and durability necessary for a safe and comfortable biking
experience. The prosthetic leg will be designed distinctly for the left and right legs based on
the application of the foot while riding the bike. With insights in prosthetics and
biomechanics, the active prosthetic ankle and adaptive equipment developed for bike riding
and locomotion in lower limb amputees will help them to have an independent lifestyle,
ultimately improving the accessibility of biking for lower limb amputees.

Keywords — Aluminum, Microcellular rubber, Hosiery stockinette, Gear motor, Bluetooth

Module, Touch Module.

144 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

AUTOMATIC HAND GLOVES THERAPY SYSTEM

S.VIGNESHWARAN,KANISHKAA.S,KEERTHANA.S,SRI DHARSHNI.S
BIOMEDICAL ENGINEERING,SRI RAMAKRISHNA ENGINEERING COLLEGE

ABSTRACT:
In recent years, there has been a rise of interest in inventing ground breaking
healthcare technologies to improve patient outcomes while reducing stress on healthcare
providers. One such focus is on developing automatic hand glove therapy devices that can be
employed in a range of rehabilitation and therapeutic applications. These novel glove therapy
devices are intended to meet specific demands, such as boosting blood circulation in impaired
people. These systems attempt to promote therapeutic activities that improve mobility and
total hand function by combining gloves with the capacity to open and close fingers. The
usage of stepper motors is the main component that allows for finger movement in these
gloves. Stepper motors are well-known for their precision control and torque generation,
making them excellent for applications that require controlled and repeatable movements. In
this case, the torque of the stepper motor is used to generate the appropriate finger
movements within the gloves, thereby improving blood circulation and providing therapeutic
advantages to users. These automatic hand glove therapy systems incorporate advanced
electronic components such as the ESP8266 microprocessor to improve their performance
and communication. The ESP8266 is a low-cost Wi-Fi microcontroller with inbuilt TCP/IP
networking software, providing both microcontroller and wireless connectivity. The gloves
can work even when there isn't direct human supervision because of a timed mechanism. The
development of a web application that benefits patients and healthcare providers is made
possible by the inclusion of Wi-Fi connectivity. Using a web application, users can remotely
track the progress of therapy, adjust settings as needed, and check real-time information on
patient performance during sessions. This ability to monitor remotely not only makes
patients' lives easier, but it also facilitates better patient-provider contact.

145 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

AI-DRIVEN EXPLORATION OF PATIENT RECORDS FOR

IMPROVED DIAGNOSIS.

Manoj Kumar B,Varniha S, Kanishq B

Computer Science and Engineering,Sri Ramakrishna Engineering College

ABSTRACT :
Artificial intelligence (AI) technology integration in healthcare has demonstrated
encouraging promise in recent years to improve patient care and medical procedures. Using
Generative AI to create a comprehensive solution for medical professionals is the goal of this
research. The suggested system consists of two primary parts, powered by a Large Language
Model (LLM): a Diagnostic Assistance Chat bot and a Medical Report Summarization Agent.
The Medical Report Summarization Agent helps medical practitioners make better decisions
by making it easier to retrieve important information from long medical reports. The module
analyses uploaded medical files and generates concise summaries using advanced Large
Language Models, providing clinicians with quick access to essential findings, diagnoses, and
recommendations. In addition, the Diagnostic Assistance Chat bot functions as a
conversational interface that allows physicians to have lively discussions on the medical
reports that have been uploaded. By utilizing the LLM's capabilities, the chat bot deciphers
user inquiries, extracts pertinent data from report summaries, and delivers perceptive answers
instantly. By enabling seamless communication and providing healthcare professionals with
on-demand help, this feature improves the system's accessibility and usability. Additionally,
the system's capability is expanded to include image analysis capabilities, enabling medical
professionals to upload diagnostic or other medical images, including X-rays. By using
cutting-edge Vison Language Models (VLM's), the system helps with the interpretation and
diagnosis of a variety of medical issues by extracting useful information from the images.
The system enhances healthcare providers' diagnostic abilities by providing a comprehensive
approach to medical decision support through the integration of textual and visual modalities.
To sum up, the suggested system is a ground-breaking attempt to use Generative AI
technology to expedite the examination of medical reports and provide diagnostic support.
The system improves medical information processing's accessibility, accuracy, and efficiency
by utilizing LLMs, which eventually improves patient care and healthcare outcomes.

146 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

DRIP INFUSION MONITORING SYSTEM

Mr.D.Gnanaprakasam, Sachin N, Navin S S , Tharun B
ELECTRICAL AND ELECTRONICS ENGINEERING ,
DR.MAHALINGAM COLLEGE OF ENGINEERING AND TECHNOLOGY

ABSTRACT:
In order to improve the safety of patients during intravenous (IV) medication in
critical care hospitals, an intelligent saline level surveillance device idea is described in the
material. It underlines how crucial it is to monitor closely in order to stop problems like blood
reflux and air embolism. In order to enable real-time monitoring and notify healthcare
practitioners of important changes, the suggested system incorporates sensors and Internet of
Things technology. In order to measure fluid levels in the intravenous (IV) treatment bag, it
has an ultrasonic sensor. Data can be sent to a centralized database for remote surveillance by
utilizing IoT connectivity. By automating monitoring processes, the system hopes to improve
patient outcomes by resolving issues with traditional monitoring and shortages of employees.

147 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

BRAIN NEOPLASM IDENTIFICATION USING DEEP

LEARNING

1MS.R.PRAVEENA AP/CSE , 2ABINAYA K, 3GOWTHAMI E, 4LAKSHMIDEVI E,

5YASHODHA G
Department of Computer Science and Engineering,
Jai Shriram Engineering College, Tirupur.

ABSTRACT:
This study offers a powerful brain tumor detection system that makes use of
deep learning techniques, especially the VGG16 learning algorithm and Convolutional
Neural Network techniques (CNN). The ultimate objective of the recommended strategy is to
enhance the accuracy and efficiency of identifying tumors in magnetic resonance imaging
(MRI) images.. Utilizing CNN allows the model to automatically extract intricate features
from input images, while VGG16, renowned for its deep architecture, contributes to a more
intricate understanding of complex patterns. The integration of these algorithms empowers
the system to discern subtle nuances indicative of brain tumors, providing a reliable and swift
diagnostic tool for medical practitioners. Experimental results demonstrate the system's
effectiveness in achieving high accuracy rates, showcasing its potential as an invaluable asset
in early tumor detection and subsequent medical intervention.

KEYWORDS: CNN, Brain tumor, Machine learning, Medical imaging

148 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

SECURE DIAGNOSIS: AUTOMATING WIRESHARK FOR

MEDICAL DATA ENCRYPTION ANALYSIS
S.Vigneshwaren(AP), M.Swetha
BioMedical Engineering, Sri Ramakrishna Engineering College ,Coimbatore

ABSTRACT :
In today's digitized healthcare landscape, ensuring the security and privacy of medical
data is paramount. One critical aspect of safeguarding this sensitive information is
encryption. However, verifying whether medical data is appropriately encrypted can be a
daunting task, particularly in large-scale networks. This paper proposes an innovative
approach to streamline this process by automating Wireshark, a widely used network protocol
analyzer. By leveraging automation techniques, our solution facilitates the efficient detection
and analysis of encrypted medical data transmissions. Through a series of experiments and
case studies, we demonstrate the effectiveness and practicality of our approach in enhancing
data security in healthcare environments. This paper presents a significant step towards
strengthening the protection of medical data and promoting trust in digital healthcare
systems.

149 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

DESIGN AND DEVELOPMENT OF AN AUTOMATIC

WOUND DRESSING APPLICATOR
Dr. Deepa B Prabhu,Saravana Venkateshan.N, Gokul Kannan.V, Vignesh Kumar.S
Biomedical Engineering,Sri Ramakrishna Engineering College

ABSTRACT:
Accurate assessment of wound characteristics, including boundary delineation,
length, and width, is essential for effective wound management in clinical practice. Manual
inspection of wounds can be time-consuming and subjective, prompting the need for
automated techniques leveraging image processing methodologies. This paper presents an
integrated approach for comprehensive wound analysis, encompassing boundary detection,
length, and width estimation, using image processing algorithms. The proposed method
begins with preprocessing steps to enhance image quality and reduce noise, followed by
robust edge detection techniques to identify the wound boundaries accurately. Concurrently,
algorithms for length and width estimation are applied to quantify the spatial dimensions of
the wound.Quantitative metrics, including precision, recall, and F1 score, are employed to
evaluate the accuracy of boundary detection, length, and width estimation against ground
truth annotations provided by domain experts. Results demonstrate the efficacy of the
proposed method in accurately delineating wound boundaries while estimating the length and
width with high precision.In conclusion, this research presents a comprehensive approach to
wound analysis, incorporating boundary detection, length, and width estimation through
image processing techniques. The proposed methodology holds promise for enhancing
wound management practices and improving patient care outcomes in diverse healthcare
settings.

Keywords:

Wound boundary detection, Image processing, Length estimation, Width estimation, Edge
detection, Healthcare efficiency

150 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

A NOVEL DESIGN OF BLDC MOTOR WITH FUZZY LOGIC

CONTROL SYSTEM
S.MAHENDRA BHARATHI,K.ARUNGANESH
POWER ELECTRONIC AND DRIVES,
PERIYAR MANIAMMAI INSTITUTE OF SCIENCE AND TECHNOLOGY

ABSTRACT :

This research abstract offers a concise overview of a study aimed at addressing torque
ripple in Brushless DC (BLDC) permanent magnet motor drives by introducing innovative
fuzzy logic-based modulation techniques for controlling DC-link voltage and current. Torque
ripple is a significant concern in BLDC motors due to its adverse impact on system
performance, efficiency, and reliability. Conventional control methods often struggle to
effectively suppress torque ripple. In response, this study proposes a novel approach utilizing
fuzzy logic-based control schemes to precisely and adaptively modulate the DC-link voltage
and current. The method involves employing fuzzy logic controllers to dynamically adjust the
DC-link parameters in real-time, resulting in a more refined and responsive control strategy.
Fuzzy logic allows the system to consider various operational conditions and input variables,
offering an effective means of minimizing torque ripple while maintaining efficient motor
performance.

151 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

PREDICTION OF LIVER STIFFNESS USING

ELASTOGRAPHIC IMAGES FIBROSCAN
Suki Melvina S,Sruthi S, Lancy J
Biomedical Engineering,Sri Ramakrishna Engineering college

ABSTRACT :

Since the late nineteenth century, medical imaging has evolved significantly,
encompassing a variety of methods for measuring physical parameters. While liver biopsy is
regarded the gold standard for determining fibrosis stage, it does not provide information
about tissue mechanical qualities, notably stiffness, which is important for disorders such as
injuries, tumours, and fibrosis. Recent improvements have created non-invasive ultrasound
procedures known as elastography, which are a viable option for examining tissue elasticity.
Thus, the use of non-invasive technologies constitutes a substantial improvement in medical
imaging.The goal of this work was to predict liver stiffness using transient elastography
(FibroScan) combined with machine learning methods. Using FibroScan images, we used
five predictive models to evaluate liver stiffness: random forest classifier, logistic regression
classifier, decision tree classifier, support vector classifier, and k-nearest neighbours (KNN).
In addition, attribute selection methods were used to minimise dimensionality and identify
the most important characteristics for successful predictions.Our classification algorithms
based on Logistic Regression and Decision Tree obtained an outstanding 93% accuracy in
predicting liver stiffness, demonstrating their ability to accurately measure liver health.
Furthermore, the random forest, k-nearest neighbours, and support vector classifier models
had accuracies of 88%. The evaluation of these models was based on various metrics
including accuracy, precision, recall, and F1 score, providing a comprehensive understanding
of their performance. Additionally, the utilization of receiver operating characteristic (ROC)
curve analysis allowed for graphical representation of the models' performance, aiding in the
visualization of their predictive capabilities. Our study highlights the potential of machine
learning techniques, coupled with non-invasive imaging modalities like transient
elastography, in accurately predicting liver stiffness. These findings contribute to the
advancement of liver disease diagnosis and management, offering clinicians valuable tools
for early detection and intervention.

152 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

AUTOMATED CLASSIFICATION OF TEXT BASED ON DISEASE

ABISHASRI S,RAJA RAJESHWARI CHANDINI P
BIOMEDICAL ENGINEERING COLLEGE,
SRI RAMAKRISHNA ENGINEERING COLLEGE

ABSTRACT :
In recent years, the exponential growth of medical data has posed a significant
challenge in efficiently analyzing and extracting valuable information from vast repositories
of textual data. Automated classification of text based on diseases has emerged as a crucial
area of research with wide-ranging implications for healthcare, including disease diagnosis,
treatment planning, and epidemiological studies. This abstract presents an overview of the
advancements, methodologies, and applications in the domain of automated text classification
focusing on diseases.The abstract begins by outlining the significance of automated text
classification in the medical domain and its potential to revolutionize healthcare by enabling
rapid and accurate analysis of textual data. This project uses the Large Language Model for
doing the natural language processing techniques. The methodologies employed in this
project is finetuning the model for the specific multi label classification task using the recent
LLM models.Employing natural language processing (NLP) techniques, particularly word
embedding and tokenization techniques are used for processing the textual data.
Subsequently, I employed a deep learning-based Large Language Model (LLM) to leverage
the extracted information. I evaluated the model's performance using three key metrics:
precision, accuracy, and F1-score. The findings demonstrate that the fine-tuned LLM model
significantly enhances the prediction accuracy of a text classification based on disease,
leveraging prior clinical information.The results indicate a significant advancement, with the
fine-tuned model achieving an accuracy of 78% on a comprehensive medical dataset. This
project contributes to the ongoing efforts in utilizing state-of-the-art NLP techniques for
automated disease classification, facilitating more accurate and efficient healthcare data
analysis.

153 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

SAFEGUARDING ANDROID ECOSYSTEM

WITHINTELLIGENTMALWAREDEFENSE
M.Jayapriya(AssistantProfessor), S.Sahana, V.KanishaSweety,
R.Velselvi( AssistantProfessor)
Dept.ComputerScience andEngineering, P.S.R. Engineering College ,Sivakasi

ABSTRACT:
Android has cornered a sizable part of the market for portable programming. Possibly
Unsafe Application (PHA) are destructive infections. Google Play Store conveys PHA in
spite of its application biological system cleanup. This portable application utilizes
progressed AI calculations to proactively recognize and alleviate potential malware dangers
on client’s gadgets.Utilizing ongoing conduct examination and mark based recognition, the
application consistently screens gadget exercises. Giving a strong safeguard against
developing malware scenes Withaeasytouseinterfaceandnegligibleassetutilization, our answer
guarantees productive and solid portable security, offering clients a consistent experience
while shielding their gadgets from malignant programming.

IndexTerms—Androidsecurity,Malwaredefense,KNN,NLP

154 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

TWITTER - BASED ELECTRONIC HEALTH RECORDS

Mugesh.T,Akashraja.M , Dhivya.N
Biomedical Engineering,Sri Ramakrishna Engineering College

ABSTRACT:

The project integrates Twitter data on medications and side effects with electronic
health records (EHRs). Social media platforms like Twitter have become important tools for
sharing information rapidly, including healthcare experiences. This project bridges the gap
between social media and healthcare records in an innovative way. It utilizes advanced data
acquisition techniques to collect relevant tweets about medications, treatment outcomes, and
side effects from Twitter. The Apify platform generates additional tweets based on specified
keywords to create a comprehensive dataset. This data is securely stored in a database.The
database is then connected to Apache Superset for data visualization and dashboard creation.
Users can explore these intuitive dashboards to analyse medication trends, sentiment analysis,
and reported side effects from Twitter. Healthcare professionals and researchers can leverage
these visualizations to identify patterns, assess medication efficacy, and monitor public
sentiment in real-time.A web interface allows users to access and interact with the
dashboards, customizing queries and filtering results. This integration of EHRs, social media
data, and data visualization enables informed decision-making in medical research, public
health policy, and personalized healthcare interventions related to medications, and public
health outcomes.

KEYWORDS:

Electronic Health Record (EHR), Apify, Social media, Healthcare, Dashboard, Medication.

155 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

ENHANCING INDEPENDENCE VISUALLY IMPAIRED

ASSISTANCE USING YOLO TECHNOLOGY

MURUGESAN R,MANOJKUMAR TS , NAVEENKUMAR S , UVARAJA LR

COMPUTER SCIENCE AND ENGINEERING
PAAVAI COLLEGE OF ENGINEERING

ABSTRACT:

Good vision is a precious gift but unfortunately loss of vision is becoming common
now a days. To help the blind people the visual world has to be transformed into the audio
world with the potential to inform them about objects as well as their spatial locations.
Objects detected from the scene are represented by their names and converted to speech.
Everybody deserve to live independently, especially those who disabled, with the last
decades, technology gives attention to disabled to make them control their life as possible. In
this work, assistive system for blind is suggested, to let him knows what is around him, by
using YOLO for detecting objects within images quickly based on deep neural network to
make accurate detection, and Open CV under Python. The obtained results indicated the
success of the proposed model in giving blind users the capability to move around in
unfamiliar indoor outdoor environment, through a user. This project involved Artificial
intelligence to detect and analyze objects and send details via speaker In the proposed work,
we solve the existing system problem by using dark net YOLO techniques that are used to
reduce the recognize time of multi objects in less time with best time complexities.

156 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

PREDICTIVE CYBER ATTACK BY USING XGBOOST

ALGORITHM IN CPMS

Mr.K . Meyyanathan1,,Mr. V. Gowtham 2Mr.M. Sarathy 3,Mrs. A. Gokul4(Assistant

Professor)
Department of Computer Science and Engineering
Paavai College of Engineering

ABSTRACT:
EHPOBS is a powerful online- banking security system that protects your information
and accounts using cutting-e-dge encryption technique-s and clever "honey passwords."
Whe-n you log into your online banking portal, the secure- system scrambles your password
into an encrypte-d code before storing it safe-ly. This encryption process adds an extra laye-r
of protection, making it nearly impossible for hacke-rs or cybercriminals to steal your actual
password. But EHPOBS goes e-ven further with its unique "hone-y password" feature.
These- are fake passwords mixed in with re-al ones, acting as decoys to confuse and de-ter
anyone trying to gain unauthorized acce-ss to accounts. It's like having a whole group of
security guards ke-eping watch over your login information. Cyberattacks can be- incredibly
damaging, putting sensitive data, financial we-ll-being, and even the- smooth running of
essential systems at risk. That's why our te-am has developed an innovative- new
cybersecurity solution using the- XGBoost algorithm. XGBoost is a highly advanced and
efficient machine- learning method that exce-ls at analyzing and finding patterns in complex
data sets. By harne-ssing the power of XGBoost, our solution can quickly dete-ct potential
cyber threats and take- action to stop them before any harm is done-. Even the most
sophisticated hacking atte-mpts won't slip past our watchful XGBoost guardian! With this
cutting-edge technology safe-guarding critical systems and information, you can operate with
confide-nce, knowing your processes and data are- secure.

Keywords— Cyber-attacks, Critical Process Management Systems, XGBoost Algorithm,

Machine Learning, Predictive Analytics, Security

157 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

ENCRYPTED HONEY PASSWORD AUTHENTICATION FOR

ONLINE BANKING SYSTEM

KABILAN T,BASKAR M , SRIKANTH S

Department of Computer Science and Engineering
Paavai College of Engineering

ABSTRACT:

Password authentication is the most widely used authentication technique, for it is

available at a low cost and easy to deploy. Many users usually set their passwords using
familiar vocabulary for its convenience to remember. Passwords may be leaked from weak
systems. Vulnerabilities are constantly being discovered, and not all systems could be timely
patched to resist attacks, which give adversaries an opportunity to illegally access weak
systems. To overcome the vulnerabilities of password attacks, here propose a password
authentication framework that is designed for secure password storage and could be easily
integrated into existing authentication systems. In proposed framework, introduces a
comprehensive security framework integrating innovative techniques to enhance password
protection and user authentication. The approach involves the incorporation of honey words
and the implementation of the AES (Advanced Encryption Standard) algorithm for secure
password storage. The password storage mechanism employs honey words as decoy
passwords, creating a deceptive layer for potential attackers. Additionally, the honey
passwords are encrypted using AES, heightening the overall security posture. During the
login process, users are authenticated through a combination of factors, encompassing
traditional password authentication method using username and password. A critical
component of the proposed system is the server-based generation and transmission of
decryption keys during login. This key allows the client to decrypt the stored password
securely. The inclusion of honey words further fortifies the verification process, as any
attempt to access these decoy passwords triggers an alert. Upon successful verification, users
gain access to the banking application, where they can engage in secure online transactions.

158 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

DATA SECURE DEDUPLICATION AND RECOVERY BASED

ON PUBLIC KEY ENCRYPTION WITH KEYWORD SEARCH

Mr.R.MANOJKUMAR,SUDHAKAR S, DINESHKUMAR A, MUKESH G

Department of Computer Science and Engineering
Paavai College of Engineering

ABSTRACT :
The explosion of data storage demands necessitates solutions for efficient cloud
storage. This work tackles this challenge by proposing a secure de-duplication and recovery
system with keyword search capabilities for cloud environments. Public key encryption
ensures data confidentiality even on untrusted cloud servers. The system intelligently
identifies and eliminates duplicate data blocks while maintaining data privacy. Users can
leverage keyword search to efficiently locate their encrypted data. Additionally, a secure data
recovery mechanism allows authorized users to retrieve their information. This approach
offers a compelling solution for cloud storage, balancing security, efficiency, and user control
over their data.

Keywords - Secure Deduplication, Public Key Encryption, Keyword Search, Cloud

Storage, Data Recovery.

159 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

AGRICULTURE AND RURAL DEVELOPMENT:

EMPOWERING COMMUNITIES FOR SUSTAINABLE
GROWTH
Ms.Velammal .M(Assistant Professor), Mavnith G S, Selvaragavan C, Silambarasan N
Computer Science and Technology,Karpagam College of Engineering,Coimbatore.

ABSTRACT :
The agricultural sector plays a pivotal role in the socio-economic fabric of rural
communities, serving as a cornerstone for food security, employment, and economic growth.
However, agricultural sustainability and rural development face multifaceted challenges
stemming from environmental degradation, resource scarcity, socio-economic disparities, and
technological advancements. This project aims to address these challenges by proposing a
comprehensive framework that integrates sustainable agricultural practices with holistic rural
development strategies. The project begins with an in-depth analysis of the current state of
agriculture and rural communities, identifying key issues such as declining soil fertility, water
scarcity, limited access to markets, and inadequate infrastructure.

160 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

CT TRANSLATOR: A PERSONALIZED TRANSLATION

SOFTWARE
MS.M.VELAMMAL(Assistant Professor), NANDAKUMAR K, MOHAMED FAISAL K,
HARISH M
Computer science and Technology Karpagam College of Engineering Coimbatore.

ABSTRACT :
This paper presents The Translator Project is a cutting-edge initiative aimed at
overcoming language barriers through the development of an advanced translation tool. In a
world characterized by globalization and multicultural interactions, effective communication
across languages is paramount. This project employs stateof-the-art natural language
processing (NLP) techniques and machine learning algorithms to create a versatile translator
capable of accurately translating text and speech between multiple languages. The tool's user-
friendly interface and robust backend infrastructure enable seamless translation for various
purposes, including personal communication, business transactions, and academic research.
By breaking down linguistic barriers, the Translator Project aims to foster cross-cultural
understanding, facilitate global collaboration, and promote inclusivity in an interconnected
world.

161 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

TOYS AND GAMES

Kanagasabapathi K(Assistant Professor), Aravind M, Venkada Krishnan S, Vairavan.M
Computer Science and Technology Karpagam College of Engineering Coimbatore.

ABSTRACT :
The project aims to develop interactive software for toys and games, merging
physical play with digital engagement to enhance children's learning and entertainment
experiences. By leveraging modern technology such as augmented reality (AR) and sensor
integration, the software will offer immersive gameplay scenarios that stimulate creativity,
problem-solving skills, and social interaction among children. The project will focus on
designing intuitive user interfaces, creating diverse game mechanics, and ensuring
compatibility with a range of toys and devices. Through iterative development and user
testing, the software aims to deliver a seamless fusion of physical and digital play, fostering
holistic and engaging experiences for children in the digital age.

162 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

TEXCO:A COMPREHENSIVE DESKTOP SOLUTION FOR

REAL-TIME MARKET INSIGHTS, ORDER TRACKING, AND
TRADING IN THE TEXTILE SECTOR
Mr.Kanagasabapathi.K(Assistant Professor), Athish K K, Balaji S, Krishnaraj V
Computer Science and Technology Karpagam College of Engineering Coimbatore

ABSTRACT :
This paper presents a desktop application tailored for the textile sector in India, addressing
key industry needs such as real-time market information, order tracking, and trading
functionalities. The application provides users with up-to-date market insights, including
pricing dynamics, trends, and news spanning various textile segments like cotton, silk, wool,
and synthetic fibers. Additionally, it offers robust tracking features, enabling users to monitor
the production and transportation stages of their orders from vendor to customer.
Furthermore, the application serves as a trading platform, facilitating direct transactions
between producers and buyers, while also supporting the creation of online marketplaces for
showcasing and selling textile products on a global scale. This comprehensive solution aims
to streamline operations, foster transparency, and bolster competitiveness within the Indian
textile industry.

Keywords— TEXTILE SECTOR , COTTON , ONLINE MARKETPLACE , ORDER

TRACKING , MARKET INFORMATION

163 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

SMART TUTOR: PERSONALIZED LEARNING

REMEDIATION TOOL
KANAGASABAPATHI K(ASSISTANT PROFESSOR), RUTHRAN G, RAKESH R,
GOWTHAM S
COMPUTER SCIENCE AND TECHNOLOGY KARPAGAM COLLEGE OF ENGINEERING
COIMBATORE.

ABSTRACT :
This paper presents to develop an online platform that provides personalized learning
remediation and tutoring services to students by matching them with the best teacher for
specific topics. This aims to revolutionize the way students access tutoring services by
employing advanced algorithms to match students with the most suitable teachers based on
their specific learning needs and preferences. By utilizing cutting-edge technologies, this
ensures a personalized learning experience tailored to each student's requirements.

164 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

A GEOGRAPHICAL INFORMATION SYSTEM BASED

APPROACH FOR EWASTE FACILITY LOCATOR
Ms. Sakthi Priya K(Assistant Professor ), Sehnaaz I T, Naveena M, Sowmiya k
Computer Science and Technology ,Karpagam College of Engineering ,Coimbatore.

ABSTRACT :
This paper presents a novel E-waste facility locator system based on a
Geographic information system technique. The expansion of electronic devices has led to an
increase in electronic waste, presenting significant environmental and health issues. Our
approach overcomes these challenges by providing users with a convenient means to locate
nearby e-waste disposal and recycling facilities. We detail the methodology, which involves a
multi-step process beginning with data acquisition, where relevant information regarding e-
waste facilities. Results show accurate information on nearby recycling centers. Our
experimental results show that the proposed system outperforms geographical information
system technology and database management systems. The system's accuracy and robustness
make it suitable for applications.

Keywords— EWASTE LOCATOR, GEOGRAPHIC INFORMATION SYSTEM,

RECYCLING CENTERS.

165 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

REAL TIME BUS TRACKING SYSYTEM : WITH LOCATION

AND ARRIVAL , DESTINATION TIME PREDICATIONS
Mr.vijay kumar P, Nithish S, Dharani Krishna S, Dharaneeswarn SV
Computer Science and Technology, Karpagam College of Engineering ,Coimbatore.

ABSTRACT :
Our project focuses on developing a user-friendly system for real-time bus
tracking and estimating arrival times at bus stops and destinations. Using GPS technology
and data from transit authorities, our system provides a simple way for users to know where
buses are and when they'll arrive. By accessing our website ,users can see a map displaying
the current locations of buses, along with estimated arrival times at nearby bus stops and
projected reaching times at their destinations. This information is updated in real-time,
allowing commuters to plan their journeys more effectively and minimize waiting times.

Keywords— GEOLOCATOR , LOCATION DETECTION ,BUS TIMMING,

NAVIGATION , TARGETING

166 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

AN AGRICULTURE PRODUCTS EXPLORER SYSTEM

THROUGH ONLINE
Ms.M.PUSHPALATHA( Professor), MS.GOMATHI M, MS.BHUVANANAYAKI S,
MS.MOWIKA A
Department of Computer Science and Technology ,Karpagam College of Engineering
Coimbatore
ABSTRACT :
This abstract introduces a novel online shopping platform tailored specifically for the
purchase of fresh vegetables and fruits. With the increasing demand for convenient and
efficient ways to acquire quality produce, this platform aims to bridge the gap between
consumers and local farmers or suppliers. The system offers a user-friendly interface
accessible through web and mobile applications ,allowing users to browse a wide selection of
fresh produce from the comfort of their homes .Key features of the platform include a diverse
range of vegetables and fruits sourced directly from trusted suppliers, ensuring freshness and
quality. Users can conveniently search, filter, and compare products based on various criteria
such as type, price, and origin. The platform also provides detailed product descriptions,
including nutritional information and farming practices, empowering consumers to make
informed purchasing decisions. In addition to facilitating individual purchases, the platform
offers subscription-based services for regular deliveries, providing users with a hasslefree
way to maintain a consistent supply of fresh produce. Integration with secure payment
gateways ensures smooth and secure transactions, enhancing the overall shopping experience.
Furthermore, the platform incorporates advanced logistics and delivery management systems
to ensure timely and reliable order fulfillment. Overall, this online shopping platform for
fresh vegetables and fruits offers a convenient, transparent, and reliable solution for
consumers seeking high-quality produce. By promoting direct connections between
consumers and suppliers, it fosters support for local farmers and promotes sustainable
agricultural practices.

167 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

AI-ENABLED DISASTER MANAGEMENT SYSTEM :

ENABLING EFFECTIVE COMMUNICATION BETWEEN
RESOURCE AGENCIES TO SUPPORT PEOPLE DURING
CRISIS
DR.V.Rajeswari(Professor), Anand CT, Obuli Sridharan G, Gopala Krishnan S
Department of Computer Science and Technology,
Karpagam College of Engineering Coimbatore
ABSTRACT :
In the dynamic landscape of disaster relief, effective coordination is important. This
paper introduces an innovative application designed to reshape how rescue agencies
collaborate and distribute resources during emergencies. At its core lies a centralized
database, the backbone of the system, storing crucial information from agency locations to
specialized skills. Utilizing intuitive interfaces and cuttingedge GPS technology, the app
provides a visually engaging map for easy agency identification and filtering based on
specific criteria. Yet, its impact extends beyond mere mapping. The application fosters real-
time communication and resource sharing among agencies, promoting efficiency and
synergy. Robust security protocols ensure data integrity and user privacy, building trust in the
platform's reliability.

Keywords— GEOLOCATOR , TRIANGULATION METHOD, LOCATION DETECTION ,

NAVIGATION , TARGETING

168 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

ENHANCE THE QUALITY OF EDUCATION IN RURAL

AREAS IN SYSTEMATIC APPROACH
MS.Sakthipriya K(Assistant Professor), Kavya T, Mahalakshmi T, Manisha S
Computer Science andTechnoloy ,Karpagam College of Engineering

ABSTRACT :
This paper presents a development of an education in rural areas in systematic
approach. Rural areas facing significant complication in providing quality education due to
Inadequate infrastructure, Lack of Languages, Teaching methods, community involvement
like these complication in rural areas. Through these community engagement and
participatory approaches, this system gather the information about unique economic, cultural,
and infrastructural factors that influence the education. This also explore the role of
technology in the education sector for rural areas. This system aim to access quality of
education and facilitate interactive learning experience for students and teachers in rural
areas. This system gives the materials for learning the courses and applying job opportunities.
This is especially for computer science students to learn programming languages. This
system provides books relates to the materials. This system seeks to improve communication
skills and relevant knowledge in rural areas people.

169 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

DEVELOPMENT OFAN ENGLISH TO HINDI

LANGUAGE TRANSLATOR TOOL FOR COMMUNICATION
OF GOVERNMENT ORGANIZATION
Ms.Sakthipriya K(Assistant Professor), Devi G, Harini S, Yuvanna Sri R
Computer Science and Technoloy ,Karpagam College of Engineering

ABSTRACT :
This paper presents a development of an English to Hindi translator System for the
communication of government organizations .This System aims to develope an efficient
translator to enhance an interaction within government organizations for easier understanding
of their relative communication .As we all know many government organization will
preferably use Hindi language so this will be useful for their interaction for speaking
constitution .We detail the methodology for translation and authentication which preferably
includes their search history to store in secure folder .This System seeks to improve
accessability and accuracy which promote better understanding within government entities .

Keywords: English to Hindi Translation ,Government Communication ,Secure

interaction ,Accuracy.

170 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

OPTIMIZING TICKET BOOKING SYSTEMS IN THE

TRAVEL AND TOURISM INDUSTRY:A DATA DRIVEN
APPROACH
Ms.Velammal M(Assistant Professor),Manu L, Poorani R, Subiksha Dharani B
Computer Science and Technoloy ,Karpagam College of Engineering

ABSTRACT :
This paper presents a novel of Ticket Booking System based on the probability and
statistics method and proposes a data driven approach to enhance ticket booking systems,
aiming to improve customer satisfaction , increase revenue and streamline operations .The
proposed approach leverages various data sources ,including historical booking data,
customer preferences ,market trends and real-time feedback, to develop predictive models
and algorithms. This adaptive capability ensures that the Ticket Booking System remains
relevant and competitive in an ever-changing landscape. These benefits include improved
conversion rates ,reduced booking errors and enhanced customer loyalty. By embracing this
technique , businesses can unlock new opportunities for growth and innovation in the
dynamic world of travel and tourism .

Keywords— PROBABILITY AND STATISTICS , DATA DRIVEN APPROACH

,STREAMLINE OPERATIONS, PREDECTIVE MODELS AND ALGORITHM.

171 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

DISASTER MANAGEMENT OF IOT

Dr V Rajeswari(Professor), Abinaya S, Chandralekha T, Mirudhula M
Computer Science and Technology Karpagam College of Engineering Coimbatore

ABSTRACT:
This paper proposes a comprehensive system for disaster management with a focus
on risk mitigation strategies implemented both before and after a disaster occurs. The system
aims to enhance preparedness, response, and recovery efforts by leveraging technological
advancements and strategic planning.

Keywords— DISASTER MANAGEMENT, RISK MITIGATION, PREPAREDNESS,

RESONSE, RECOVERY, PRE PLANNING AND POST PLANNING OF DISASTER.

172 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

DIGITAL ASSISTANT FOR LEGAL AWARENESS AND

DESIGNING A KYR KNOW-YOUR-RIGHTS FRAMEWORK
IN INDIA
Ms.Velammal.M(Assistant Professor), Vignesh G, Someshwaran R, Seetharaman A
Computer Science and Technology, Karpagam College of Engineering, Coimbatore.

ABSTRACT:
Access to legal information and awareness is a significant challenge in India,
particularly for those who are illiterate or marginalized. This paper presents the design and
development of a user-friendly platform aimed at enhancing legal awareness among these
underserved populations. The platform leverages simple language, visual aids, and interactive
features to make legal concepts and procedures more understandable and accessible. By
focusing on user needs and usability, the platform seeks to empower individuals with legal
knowledge, thereby bridging the gap in legal literacy and facilitating access to justice. The
paper discusses the design principles, development process, and potential impact of the
platform on improving legal awareness and access to justice in India. Preliminary user testing
results indicate positive feedback and suggest the platform has the potential to significantly
enhance legal awareness and empower marginalized communities.

173 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

AI ENABLED HRBOT HELPDESK FOR EMPLOYEES

COLLABORATION AND COMMUNICATION
Mr.N.VIJAYAKUMAR(Assistant Professor), AMRT S, ABISHEK K, RAMASURYA
Computer Science and Technology, Karpagam College of Engineering, Coimbatore.

ABSTRACT:
This paper talks about a cool thing called the HRBOT Helpdesk. It’s like having a
smart robot that helps with work stuff. This robot uses fancy technology called AI to
understand what people say and help them out. It’s like having a friend who knows a lot
about work things. This HRBOT Helpdesk makesit easier for people to talk to each other and
work together. It helps answer questions fast and makes sure everyone has the right
information. This papershows how the HRBOT Helpdesk can make work more fun and help
everyone get along better. It also talks about how we can make sure the robot is fair and
works well for everyone.

174 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

ENHANCING ACADEMIC INTEGRITY THROUGH TECHNOLOGICAL

INNOVATIONS: A COMPREHENSIVE REVIEW OF PROCTORED
EXAMINATION TOOLS AND EMERGING TRENDS
Ms.M.Pushpalatha(Professor), Mathuminth S, Ram prasad S, Vishwa V
Computer Science and Technology, Karpagam College of Engineering, Coimbatore.

ABSTRACT:
This paper presents the development of a state-of-the-art proctored exam tool
designed to uphold the utmost integrity in online assessments. With the exponential growth of
remote learning, ensuring the authenticity of exams has become paramount. Our tool
integrates cutting-edge technology, including facial recognition and anti-cheating algorithms,
to authenticate test-takers and monitor their activity in real-time. By providing a secure
testing environment and deterring dishonest behaviors, our tool safeguards the credibility of
online exams. This paper details the features and implementation of our proctored exam tool,
offering a robust solution for educational institutions and organizations seeking to maintain
the integrity of remote assessments.

175 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

ADVANCES IN REAL TIME MONITORING OF

AGRICULTURAL DEVELOPMENT: ENHANCING
EFFICIENCY SUSTAINABILITY AND RESILIENCE
DR.PUSPALATHA(Professor), Sibi Chandar R, Praveen Kumar S, Karthic E
Computer Science and Technology, Karpagam College of Engineering, Coimbatore.

ABSTRACT:
(RTM) Real-time monitoring has emerged as a pivotal tool in revolutionizing
agricultural development, offering timely insights and actionable data to farmers, researchers,
and policymakers. This paper explores the significance of real-time monitoring systems in
enhancing agricultural productivity, sustainability, and resilience. By integrating technologies
such as Internet of Things (IoT), remote sensing, and data analytics, real-time monitoring
enables continuous tracking of crucial parameters including soil moisture levels, weather
conditions, crop health, and pest infestations. The utilization of such systems facilitates
informed decision-making, precise resource allocation, and proactive management practices.
Furthermore, real-time monitoring fosters adaptive strategies to mitigate risks associated with
climate change and market fluctuations, thereby bolstering the resilience of agricultural
systems. This paper also discusses the challenges and opportunities in implementing real-time
monitoring solutions, emphasizing the need for interdisciplinary collaboration, scalable
infrastructure, and user-friendly interfaces. Through case studies and examples, the paper
illustrates the transformative potential of real-time monitoring in driving sustainable
agricultural development across diverse contexts. Overall, real-time monitoring represents a
powerful approach to optimize agricultural practices, improve yield stability, and contribute
to global food security.

Keywords: Real-time monitoring, Agricultural development, Sustainability, Internet of

Things (IoT), Remote sensing, Data analytics, Precision agriculture, Climate resilience,
Decision support systems, Interdisciplinary collaboration.

176 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

ADVANCING DOMESTIC WASTE MANAGEMENT

THROUGH SMART TECHNOLOGIES AND SUSTAINABLE
PRACTICES
Mr.N.VijayaKumar, Manoj Kumar D, Prasath S, Surya .K
Computer Science and Technology, Karpagam College of Engineering, Coimbatore.

ABSTRACT:
In the dynamic landscape of Efficient management of domestic waste is essential for
maintaining environmental sustainability and public health.the design and implementation of
an integrated domestic waste management system aimed at optimizing waste handling
processes while promoting recycling and minimizing environmental impact. The proposed
system integrates several key components to streamline domestic waste management. Firstly,
it includes a comprehensive waste segregation mechanism implemented at the source to
encourage residents to separate recyclable materials, organic waste, and non-recyclables. This
segregation is facilitated through the distribution of color-coded bins and educational
campaigns to raise awareness about the importance of waste separation

Keywords-WASTE SEGREGATION,RECYCLING,WASTE COLLECTION,WASTE

DISPOSAL,COMPOSTING

177 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

VIRTUAL ZOO APPLICATION: LIVE ANIMAL CAMS

MR.VIJAYAKUMAR N(ASSISTANT PROFESSOR), SOMESH P, VISHNUPRASATH,
ANBUNESAN S
Computer Science and Technology, Karpagam College of Engineering, Coimbatore.

ABSTRACT:
This paper presents The Virtual Zoo: A Holographic Display System for Immersive
Education on Flora, Fauna, and Ecosystems .As humanity faces increasing challenges related
to environmental conservation and education, innovative approaches are necessary to inspire
and educate future generations about the importance of biodiversity and ecosystem
preservation. In response to this need, we present the concept of the Virtual Zoo, a
groundbreaking holographic display system designed to showcase both existing and extinct
species of flora and fauna within their natural ecosystems.The Virtual Zoo integrates
advanced hardware and software technologies to create immersive educational experiences
that can be displayed in schools, museums, and government institutions without requiring
extensive physical space. By leveraging holographic projection technology, viewers are
transported into vivid depictions of diverse ecosystems from around the world, allowing them
to explore and learn about the intricate relationships between species and their environments.

178 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

EMPOWERYOUNGRIGHTS: DEVELOPMENT OF
GAMIFIED PLATFORM ON CHILDREN'S RIGHTS TO
INCREASE LEGAL LITERACY AND AWARENESS AMONG
CHILDREN IN INDIA
PUSHPALATHA M(ASSISTANT PROFESSOR), ALEXSHA PON CHALES M,
JAMUNADHARSHINI V, YAZHINI S
Computer Science and Technology, Karpagam College of Engineering, Coimbatore.

ABSTRACT:
This paper presents to develop an gamified platform to educate children in
India about their rights that seeks to make learning rights engaging, interactive and accessible
to children across different age groups. This aims to enhance legal literacy and raise
awareness among children in India about their rights particularly focusing on children’s rights
by employing gamification techniques. By harnessing the transformative power of
gamification and digital technology, this platform paves the way for a more inclusive,
equitable, and rights-respecting society, where every voice is heard and every right is upheld.

179 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

COMMUNITY BASED REPORTING AND MONITORING

TOOL FOR WOMEN’S SAFETY IN
COLLEGES/UNIVERSITIES
Dr.Rajeswari V(Professor), Oviya D K, Shraya S
Computer Science and Technology ,Karpagam College of Engineering ,Coimbatore.

ABSTRACT:

This paper introduces a community-driven approach to addressing the issue

of women's safety in colleges and universities through the development of an innovative
reporting and monitoring system. The system leverages modern technology to provide a user-
friendly platform for women to report incidents of harassment, assault, or other safety
concerns in real-time. By fostering a sense of community and solidarity, the system
encourages active participation from students, faculty, and staff in promoting a safer campus
environment. Furthermore, the system facilitates data collection and analysis to identify
trends and patterns, enabling targeted interventions and policy enhancements. Through
collaboration with stakeholders and the implementation of proactive measures, this system
aims to empower women and create a culture of safety and respect in higher education
institutions.

180 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

A UNIFIED PLATFORM EMPOWERING CAREERS WITH

REACT AND DJANGO INTEGRATION
Ms.M.Pushpalatha, Prasannakumar M, Manoj Kumar E, Ranjith R
Computer Science and Technology ,Karpagam College of Engineering ,Coimbatore.

ABSTRACT:

The National Web Portal for Job-Oriented Courses serves as a pivotal platform for
aligning skill development initiatives with industry demands. Leveraging job descriptions
provided by government officials, the portal facilitates the creation of courses tailored to meet
the evolving needs of the workforce.Through advanced extraction and analysis techniques,
job descriptions are processed to identify key information such as required skills,
qualifications, and responsibilities. Users interact with the portal through an intuitive
interface, allowing for seamless navigation and collaboration among job seekers, course
designers, and administrators.

181 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

AI-BASED IT TRAINING SYSTEM: FACILITATING

EFFICIENT COMMUNICATION AMONG TRAINING
RESOURCES TO EMPOWER LEARNERS IN SKILL
DEVELOPMENT
Dr.V.Rajeswari, Dhuruvan Shanker R, Srimoulidharan M, Suryaprakash M
Computer Science and Technology ,Karpagam College of Engineering ,Coimbatore.

ABSTRACT:

This paper introduces a revolutionary system reshaping the landscape of IT training.By

harnessing the power of AI Technology,it facilities seamless collaboration among training
institutions while delivering personalized learning experiences to students.The system serves
as a centralized hub for storing essential course materials and tracking student
progress.Through user-friendly interfaces and intelligent AI algorithms,it customizes learning
to suit individual learning styles.It provides instant feedback and adapts its teaching methods
based on student performance.

182 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

ONLINE XEROX AUTOMATION

Tharun Akhash A, Shree Guhan K, Thanga Kumar M
Computer Science And Engineering, Sri Ramakrishna Engineering College

ABSTRACT:

This paper investigates the application of online printing Xerox automated

services in educational institutions like schools and colleges which are useful for students and
faculties.This application covers the part of consumption of energy spend by the people in
taking printouts which makes it easier for everyone to complete their work before due date or
deadlines.

183 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

DRIVER DROWSINESS DETECTION SYSTEM

Mrs.D.Rajeswari, Pavilaa.C, Abitha.R, SaiVarshini.D
Department of computer science and engineering
Avinashilingam Institute for Home Science and Higher Education for Women

ABSTRACT:

In the contemporary era of burgeoning road accidents attributed to driver drowsiness

and distraction, the imperative to develop advanced safety measures has never been more
pronounced. This project endeavors to introduce a pioneering solution, harnessing the power
of OpenCV and image processing techniques to monitor a driver's facial and eye movements
in real-time. By meticulously analyzing crucial parameters such as blink rate and head pose,
the system can promptly identify signs of drowsiness or distraction, triggering audible alerts
and notifying relevant authorities. This multifaceted approach aims to significantly enhance
road safety, mitigate the risk of accidents, and revolutionize the landscape of driver
monitoring systems.

Keywords: Drowsiness detection, Face detection and Tracking, Eyes Detection, OpenCV

184 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

DEEPFAKE DETECTION OF SYNTHETIC MEDIA USING

LSTM ALGORITHM
MRS.S.BHARANI NAYAGI,THARUN PV, KARTHICK P, G.S.MITHUN KRISHNA
DEPARTMENT OF ARTIFICIAL INTELLIGENCE AND MACHINE LEARNING
M KUMARASAMY COLLEGE OF ENGINEERING

ABSTRACT:

Deepfake technology has emerged as a significant threat to the integrity of

digital media, prompting the need for robust detection methods. In this study, we propose a
DeepFake Detection system utilizing Long Short-Term Memory (LSTM) algorithms.
Leveraging the temporal dynamics inherent in video sequences, our approach aims to
accurately distinguish between authentic and synthetic media. We collected a diverse dataset
encompassing both genuine and deepfake videos for training and evaluation. Through
rigorous experimentation, our LSTM-based model achieved a remarkable accuracy of 95%,
demonstrating its efficacy in detecting deepfake content. The precision, recall, and F1 score
were 96%, 94%, and 95% respectively, underscoring the model's ability to minimize false
positives while capturing true positives. Our study contributes to the advancement of
deepfake detection methodologies, highlighting the potential of LSTM algorithms in
combating the proliferation of synthetic media.

Keywords—DeepFake Detection, Synthetic Media, Long Short-Term Memory (LSTM),

Temporal Dynamics, Video Analysis, Accuracy, Precision, Recall, F1 Score..

185 ICI3’2K24
PROCEEDINGS OF INTERNATIONAL CONFERENCE ON INNOVATION IN INTELLIGENCE AND INFORMATICS
ISBN-NUMBER : 978-81-971984-9-6

ANTISLEEP ALARM PROJECT –A PROBLEM SOLVING

IDEA FOR DRIVERS
P.Balamurugan,Panimalar Engineering College
G.J.Aparnaa,Kamaraj College of Engineering College

ABSTRACT:

Imagine a situation where a person is driving home from work,dead tired after facing
all the challenges of the day.His hands are on the wheel and foot on the pedal but suddenly he
starts feeling drowsy,his eyes start shutting and his vison blurs and before he knows it,he’s
asleep.Falling asleep on the wheel can lead to serious consequences,there may be accidents
and people may even lose their lives.This situation is much more common then we notice and
hence,it is very important to counter this problem.So to address this issue,we have come up
with a Driver Anti-sleep device.This system alerts the user if he/she falls asleep at the wheel
thereby,avoiding accidents and saving lives.

186 ICI3’2K24