Proceedings of the Third International Conference on Electronics Communication and Aerospace Technology [ICECA 2019]

IEEE Conference Record # 45616; IEEE Xplore ISBN: 978-1-7281-0167-5

AI and Web-Based Human-Like Interactive

University Chatbot (UNIBOT)
Neelkumar P. Patel1 Prof. Darshan A. Patel3
Devangi R. Parikh2 Prof. Ronak R. Patel4
Department of Information and Technology Department of Information and Technology
Sardar Patel College of Engineering Sardar Patel College of Engineering
Bakrol, Gujarat 388315 Bakrol, Gujarat 388315
Email: [email protected] , Email: [email protected] ,
[email protected] [email protected]

Abstract—Most of the time, Students have to visit universities ALICE bot which uses AIML(Artificial Intelligence Mark-up
or colleges to collect various information like Tution fees, Term Language) and program Eliza. Such chatbot performs pattern
Schedule, etc. during their admission process or as per their matching which requires particular patterns to be matched [1]
daily needs. This process is very tedious and time consuming,
also it requires manpower in providing required information [2]. Hence, a chatbot named “UNIBOT” is developed. It deliv-
to visitors. Hence, to overcome the problems a chatbot can be ers efficient and relevant response to the user corresponding to
developed. The project is about interaction between users and their entered message. The interface is effectively interactive.
chatbot which can be accessed from anywhere anytime. The The time of response is minimal. It requires less memory and
chatbot can be easily attached with any university or college database hits are very less. Majority of the time, the provided
website with few simple language conversions. Chatbot provides
various information related to university or college and also reply satisfies the user’s requirement. The algorithm defined
students-related information. The chatbot can be used by anyone in this paper can also be used as back-end to develop android
who can access the university’s website. The project uses the chatbot applications like [10] [11] or in health care areas [12]
concept of Artificial Intelligence and Machine Learning. PHP for faster and efficient response.
Language is utilized for the development of Chatbot. User can
ask university-related questions, then the query is applied as an II. P ROPOSED S YSTEM
input to algorithm, which processes the message and displays the
corresponding response to the user. The Project GUI is similar
to a Messaging Application.

Keywords—Artificial Intelligence; Chatbot; Human-like interac-

tive; Machine Learning; University Chatbot

I. I NTRODUCTION
The project deals with user’s request in form of question-
based message and processes it to deliver a desired response
in form of message. It solves the process of visiting colleges
and gathering related information as per the needs, as it is
time consuming. Also, the user can communicate to admin
office with telephone number provided but doesn’t receive
a positive feedback. The project is a web-based chatbot.
Graphical User Interface (GUI) is much similar to messaging
application, which provides a friendly environment to the user
as they are much aware of operating messaging applications.
The user types a question and on performing submit, the
message is preprocessed and the most relevant information
from the database is provided as a response in similar way of
messaging. Developing a chatbot solves the problems that can
arouse in gathering required information. It can be accessed
from anywhere at anytime. In various websites, users are not Fig. 1: Flowchart
able to find the required information on website which in turn
end up closing the websites, which can be fulfilled by using The flowchart of the system displays how the Unibot
chatbot. Presently, there are various chatbots available like performs. Initially, the user message is preprocessed and con-

978-1-7281-0167-5/19/$31.00 ©2019 IEEE 148

 Proceedings of the Third International Conference on Electronics Communication and Aerospace Technology [ICECA 2019]
IEEE Conference Record # 45616; IEEE Xplore ISBN: 978-1-7281-0167-5

nectivity to database is obtained. Then, based upon conditions B. Algorithm Implemented

satisfied the Unibot process flows and provide response to user. A totally new algorithm is developed & implemented in
PHP language is used as a back-end to achieve the represented Unibot. It is very efficient, requires less memory and has
scenario. minimal database hits. The algorithm is as follows:
1. Accept the message from the user.
III. D ESIGN OF U NIBOT 2. Now, perform the following to the accepted message:
• Firstly, check each word whether it is spelled cor-
The chatbot is also known as “UNIBOT” i.e “University rectly or not. If not, show suggestions to the user as
Chatbot”, The design is shown in Figure 2. Graphical User shown in Fig. 4 and perform operation based upon
Interface (GUI) is an important component of any system. The selected option.
front-end is developed using HTML, CSS and jQuery. Ajax is – If Yes is selected, perform the following opera-
used to call and get response from PHP file, whereas, jQuery tion using the suggested string.
is used to display the messages to the user. – Otherwise, perform the operations using the
user’s previous string.
• Split the message into words.
• Execute an SQL using Regular Expression to check
the words are available in the database.
• Store the words present in the database into an array
called “important words”.
• Execute SQL query using the above array words.
• If the result of above step produces single row then
display the answer to the user.
• If multiple rows are produced, then display options
to the user with the help of “title” column in the
table.
Fig. 2: Design of Unibot • If no result is produce, then check all the keywords
in “Answer” columns. Do the following:
IV. I MPLEMENTATION P ROCESS – If a match is found, store the user’s question
and found answer into another table. So, if same
The implementation of Unibot was performed by the fol- question is asked, the chatbot can provide answer.
lowing: – Otherwise, display sorry message to user.
• Based on options selected or another message en-

A. Database Design tered, Go to step 2.

V. S AMPLE R ESULTS
The database is the most important part of any system. The
table is created as below and the data is entered in such a way Static chatbots only involves comparing of strings, whereas
that if a question has more than one answers, then they are Unibot uses dynamic approach like preprocessing the message
stored in Answer 2 and Answer 3 fields. Also, the Question before searching for a response [4] [5].
only contains keywords rather than particular natural language
question as shown in Table I. Unibot provides details such as
departmental syllabus, events, admission procedure and fees,
basic university details, class timetable, important circulars,
etc [3].

TABLE I: Question and Answer table

QA ID Title Question Answer 1 Answer 2 Answer 3

How
Hey may
1 hello Heyy
there. I help
you?
Happy
Good See you
2 bye to help (a) Message Received (b) Options Displayed
Bye. again.
you.
Fig. 3: Variant responses
spce
Rs.
Yearly fees
3 60,000
Fees college
per year The user enters the message in the text field and submit
yearly it. The message is instantly displayed in the chat window

978-1-7281-0167-5/19/$31.00 ©2019 IEEE 149

 Proceedings of the Third International Conference on Electronics Communication and Aerospace Technology [ICECA 2019]
IEEE Conference Record # 45616; IEEE Xplore ISBN: 978-1-7281-0167-5

with the help of jQuery. The user’s question is preprocessed information by visiting websites or colleges.
through which irrelevant words are neglected, then connection
with database is performed to search for the most relevant
VII. F UTURE S COPE
information. The response is shown in the chat window in
the form of message received as in messaging application on The project can be integrated with any university or college
mobile devices. Majority of chatbots provides default answer website. Along with educational information, the chatbot can
when no match is found [6] [7]. Whereas, Unibot displays be extended by providing other relevant information. It will be
related options as shown in Fig. 3b. helpful to students as well as other visitors. Thus, the chatbot
can provide a wide range of information as per developer’s
configuration. Natural Language Processing (NLP) can also
be integrated to enhance the chatbot.
R EFERENCES
[1] Drishti Malik Vibhor Sharma, Monika Goyal, “An intelligent behaviour
shown by chatbot system”, International Journal of New Technology and
Research, 2017.
[2] Balbir Singh Bani, Ajay Pratap Singh, “College Enquiry Chatbot Using
A.L.I.C.E”, International Journal of New Technology and Research 2017.
[3] Nitesh Thakur, Akshay Hiwrale, Sourabh Selote, Abhijeet Shinde, Prof.
Namrata Mahakalkar, “Artificially Intelligent Chatbot”, Department of
Computer Science & Engineering, Priyadarshini Institute of Engineering
& Technology, Nagpur.
[4] M. Dahiya, “A tool of conversation: Chatbot”, Dept. of Computer Science,
(a) When NO is selected (b) When YES is selected Maharaja Surajmal Institute, Janakpuri, India, 2017.
[5] Tracy A. Tindall Michael G. Makar, “Dynamic chatbot”.
Fig. 4: Suggestions [6] B. R. Ranoliya, N. Raghuwanshi and S. Singh, “Chatbot for University
Related FAQs”, 2017 International Conference on Advances in Comput-
ing, Communications and Informatics (ICACCI), Udupi, 2017.
Analysis of [8] aided to think about providing suggestions [7] Amey Tiwari, Rahul Talekar, Prof.S.M.Patil, “College Information Chat
to user. If the user misspells words in the message submitted, Bot System”, International Journal of Engineering Research and General
Science 2017.
the chatbot also provides suggestions to the user as shown in [8] Himanshu Agarwal, Rameshwari Deshmukh, Supriya Biradar, Bhagyashri
Fig. 4. Based on options selected, corresponding response is Cheke, “Chat-Bot Real Virtual Human”, International Journal of Ad-
displayed. vanced Research in Computer and Communication Engineering 2018.
[9] Anirudh Khanna, Bishwajeet Pandey, Kushagra Vashishta, Kartik Kalia,
Bhale Pradeepkumar and Teerath Das, “A Study of Today’s A.I. through
Chatbots and Rediscovery of Machine Intelligence”, International Journal
of u- and e- Service, Science and Technology 2015.
[10] Sarthak V. Doshi, Suprabha B. Pawar, Akshay G. Shelar, Shraddha
S. Kulkarni, “Artificial Intelligence Chatbot in Android System using
Open Source Program-O”, International Journal of Advanced Research
in Computer and Communication Engineering, 2017.
[11] Fernando A. Mikic Fonte, Martı́n Llamas Nistal, Juan C. Burguillo
Rial, Manuel Caeiro Rodrı́guez, “NLAST: A natural language assistant
for students”, Department of Telematics Engineering, University of Vigo,
Vigo, Spain.
[12] V.Manoj Kumar, A.Keerthana, M.Madhumitha, S.Valliammai,
V.Vinithasri, “Sanative Chatbot For Health Seekers”, International
Journal Of Engineering And Computer Science 2016.

Fig. 5: User’s details

Unibot also can save user’s details, if the user write query in
the format “my..is..”. This details are saved in local storage of
user’s computer using jQuery [9]. Hence, it will remain until
user deletes all it’s caches. Whenever user ask about his/her
details, Unibot provides correct answer as shown in Fig. 5
VI. C ONCLUSION
A chatbot is the best tool which provides quick way to
interact with the users. It is very helpful to the users as it
allows to enter questions in natural language and desired
information is obtained easily to the user. In this paper,
details about design, algorithm used and implementation of
the Unibot is presented. The user doesn’t need to gather

978-1-7281-0167-5/19/$31.00 ©2019 IEEE 150