This may be the author’s version of a work that was submitted/accepted

for publication in the following source:

Karthick, S., Victor, R. John, Manikandan, S., & Goswami, Bhargavi

(2018)
Professional chat application based on natural language processing.
In Proceedings of the 2018 IEEE International Conference on Current
Trends in Advanced Computing (ICCTAC 2018).
Institute of Electrical and Electronics Engineers Inc., United States of
America, pp. 1-4.

This file was downloaded from: https://eprints.qut.edu.au/197780/

c IEEE

This work is covered by copyright. Unless the document is being made available under a
Creative Commons Licence, you must assume that re-use is limited to personal use and
that permission from the copyright owner must be obtained for all other uses. If the docu-
ment is available under a Creative Commons License (or other specified license) then refer
to the Licence for details of permitted re-use. It is a condition of access that users recog-
nise and abide by the legal requirements associated with these rights. If you believe that
this work infringes copyright please provide details by email to [email protected]

Notice: Please note that this document may not be the Version of Record
(i.e. published version) of the work. Author manuscript versions (as Sub-
mitted for peer review or as Accepted for publication after peer review) can
be identified by an absence of publisher branding and/or typeset appear-
ance. If there is any doubt, please refer to the published source.

https://doi.org/10.1109/ICCTAC.2018.8370395
 Professional Chat Application based on Natural
Language Processing
Karthick S1, R John Victor2, Manikandan S3, Bhargavi Goswami4
Department of Computer Science, Christ University,
Bengaluru, Karnataka, India.
Email: [email protected], [email protected],
[email protected], [email protected]

Abstract— There has been an emerging trend of a vast proposed mechanism implemented. The proposed network-
number of chat applications which are present in the recent years based android chat application used for chatting purpose with
to help people to connect with each other across different remote clients or users connected to the internet, and it will
mediums, like Hike, WhatsApp, Telegram, etc. The proposed not let the user send inappropriate messages. The application
network-based android chat application used for chatting developed for Android, because Android is one of the most
purpose with remote clients or users connected to the internet, widely used mobile operating systems having major market
and it will not let the user send inappropriate messages. This share when compared to other mobile operating systems like
paper proposes the mechanism of creating professional chat iOS, Windows and Blackberry [9,10].
application that will not permit the user to send inappropriate or
improper messages to the participants by incorporating base
level implementation of natural language processing (NLP).
Before sending the messages to the user, the typed message The rest of the paper organized as Section II approaches
evaluated to find any inappropriate terms in the message that related work. Section III outlines the proposed methodology.
may include vulgar words, etc., using natural language Section IV explains implementation and conclusion of the
processing. The user can build an own dictionary which contains paper.
vulgar or irrelevant terms. After pre-processing steps of removal
of punctuations, numbers, conversion of text to lower case and II. RELATED WORK
NLP concepts of removing stop words, stemming, tokenization,
named entity recognition and parts of speech tagging, it gives A. Enhanced Education Chat Application Based Chat
keywords from the user typed message. These derived keywords
Application Based on Interested Keyword with Username
compared with the terms in the dictionary to analyze the
sentiment of the message. If the context of the message is and Password Authentication Security
negative, then the user not permitted to send the message. This paper states that since there is a lot of changes which
had occurred in the recent past and the customers are changing
Keywords— Android Application, Chatting, Dictionary, Named their needs from time to time, it has become a necessity to
Entity Recognition, Natural Language Processing, Networking, build a chat application that matches the present crowd. This
Parts of Speech tagging, Sentimental Analysis, Stemming, system was mainly developed to provide a chat room for all
Tokenization. kind of educationalist by offering them a platform to chat and
solve study related queries [11].
I. INTRODUCTION
Online chatting refers to the process of sending and
receiving messages using the internet. There are various B. Design of Chatting Application Based on Android
chatting applications available in the market. At the first Bluetooth
quarter of 2017, the total number of users using chat
applications are more than 5.03 Billion [1], and widely used This paper proposes a method of sending and receiving
apps are WhatsApp, Facebook Messenger, We Chat, QQ text messages using the Bluetooth connection. Since the
Mobile, etc., All these applications provide various features to Bluetooth technology consumes low cost and power, they
ensure security, integrity, and consistency. All these apps let incorporated a system that enables the user to chat with each
the user send any messages, and the messages can be lewd or other over a Bluetooth connection. But this method might also
inappropriate. There are many cases filed for sending lewd or face constraint as the range of the Bluetooth connection is
inappropriate messages in various online mediums [2,3,4,5,6]. restricted [12].
It may also be possible for the user to send inappropriate
messages by mistake. According to Section 66A of the
Information Technology (Amendment) Act, 2008 says that C. Android forensics analysis: Private chat on social
transmitting of obscene information using a transmission messenger
equipment which may result in three years of incarceration This paper discusses regarding the relic data from secret,
including a fine [7,8]. In order to solve these concerns, the hidden and private chat. They also provide reports of formed
messages along with how it associates to one another. From context in the message and prevents the user from sending
the inquiry outcomes of Android forensics and interpretation, those messages with a warning notification.
an inquisitor or investigator will be able to understand,
reproduce, and confer the chronology of the information
which has been generated by the user [13]. III. PROPOSED METHOD
Failure to identify the inappropriate context in the text
message is the main reason for various problems. The
D. Extracting Intrauterine Device Usage from Clinical Texts proposed system solves these issues by developing an NLP
using Natural Language Processing based android tool which identifies and warns the user if the
This paper discusses the devices are much useful regarding sentiment of the message contains lewd or vulgar context. The
preventing pregnancy which was not intended to happen. The proposed system illustrated in Fig.1 consists of three phases.
Clinical texts help the patient to overcome this risk factor. But
after a series of research, the results showed that these data are
structured and cannot be efficiently used for processing and
analyzing the risk factor but would be used in the traditional
method of maintaining health record. To overcome this
problem, a clinical factor extraction tool used called EasyCIE
which was used to identify the patients who are contraception
for counseling [14].

E. Dialogue-Oriented Interface for Linguistic Human-

Computer Interaction: A Chat-based Application
This paper discusses a model allowing human-computer
interaction by utilizing natural language utterance.
Development of interactive agents like chatbots, in the
recommended environment, the human-machine discussion is
dedicated to a question or answer monotonic approach
proposed at reducing semantic uncertainty within information.
The prevalence of the chat as asynchronous transmission
device permits design be fit for an infinite diversity of forms
[15].

F. Lexical and Discourse Analysis of Online Chat Dialog

This paper discusses one of the final goals of natural
language processing (NLP) methods is getting the definition
of what is transpiring sent, irrespective of the means (e.g.,
printed versus lectured) or the pattern (e.g., static reports
versus productive arguments). Although significant work has
transpired performed in conventional language domains such
as speech and static written text, understanding has yet been
done in the latest information areas approved by the Internet,
e.g., online chat and immediate messaging. That is in part due
to the point that there are no explained chat corpora accessible Fig. 1. Block Diagram of the Proposed System
to the broader research area. This dissection aims to develop a
chat corpus, marked with lexical (part-of-speech token labels), The first phase deals with data pre-processing. The result of
syntactic, and speech data. Such a corpus can be practiced to the first step given as an input to the second phase. The second
improve more complicated, analytical-based NLP forms that phase deals with implementation of NLP concepts like the
do tasks such as artist profiling, item description, and social removal of stop words, stemming, entity recognition,
network reports [16]. tokenization and parts of speech tagging which derive
keywords from the user typed the message. These keywords
The related work states that there are no methods or provisions compared with user dictionary to identify irrelevant terms. The
for identification of lewd or vulgar context in the typed third phase deals with sending and receiving the messages
message of a user and stop the user from sending the message. using the internet and saving the messages in encrypted form
The proposed methodology solves these issues by developing in the real-time database "FireBase".
an NLP based android tool for identification of lewd or vulgar
 IV. IMPLEMENTATION limited importance in capturing the semantics of the text and
The NLP based tool implemented using Android, Java, and additionally stop words increase the searching time which
Firebase. The tool accepts the user typed message as an input. leads to wastage of many computational resources. Stop words
Data pre-processing applied to the user typed message. are omitted to preserve both time complexity and space
complexity. Stemming is a crude method that cuts off the ends
or beginning of words. The primary aspiration of stemming is
to change a derivative word into its standard form and keeps
the root word. Parts of Speech Tagging examines the text and
assigns parts of speech to each token as a verb, adjective,
noun, etc., Entity Identification helps to classify the named
entities like persons, organizations, etc., from the text. After
implementing NLP techniques on the pre-processed cleaned
data, definitive keywords derived. These keywords compared
with the keywords available in the user dictionary. The
vocabulary contains all the keywords. From the keywords
procured from the text, the sentiment of the context
determined. If the meaning of the message is inappropriate,
then user not allowed to send the message.

Fig. 2. Splash Screen of Professional Chat Application

Data pre-processing is a procedure of data mining, executed

on real-time data, which is vociferous, inadequate or erratic.
The subsequent data scrubbing measures are employed:
(i) Eliminating all the distinctive letters from the line.
(ii) Trimming undesired tabs, spaces, newlines and
additional nonprintable letters in the line.
(iii) Transforming the entire text to lower case.
All these steps performed to make computation easy.

Natural Language processing concepts applied to the pre-

processed data. Natural Language Processing is a method for Fig. 3. Displaying warning message if the user tries to send inappropriate
machines to evaluate, determine, and obtain the semantics of message
human language wisely. Java Programming language is
employed to accomplish several tasks of Natural Language
Processing on cleaned data. Word Tokenization, Sentence The information stored in a secured real-time database
Tokenization, Removal of Stop words, Stemming, Entity Firebase [17] [18]. The data stored in a JSON file which is
Recognition, Parts of Speech Tagging, etc. are practiced to easy to process. The MD5 algorithm used to convert the user
reconstruct the data to a pattern suitable for interpretation. typed message into fixed size hash of 32 digits’ hexadecimal
Word Tokenization is the method of changing the text into value [19].
tokens and saving it in the list. Sentence tokenization is the
method of transforming the text into different phrases. Stop
words deemed inapplicable or pointless because they have
 [online] http://ijlljs.in. Available at: http://ijlljs.in/wp-
content/uploads/2015/06/CASEANALYSISFORIJLLJS.pdf.
[4] Citizen Matters, Bengaluru. (2014). How not to get arrested under 66A
for your online chatter |. [online] Available at:
http://bengaluru.citizenmatters.in/how-not-to-get-arrested-under-66a-
for-your-online-chattering-6589.
[5] Mhaiske, S. (2017). Abusive Chat. [online] Cybercrimcomplaints.
Available at:
https://www.cybercrimecomplaints.com/category/complaint-
types/abusive-chat.
[6] The Hindu Business Line. (2015). Man arrested for sending lewd SMS
to lady prof denied bail. [online] Available at:
http://www.thehindubusinessline.com/news/national/man-arrested-for-
sending-lewd-sms-to-lady-prof-denied-bail/article4512487.ece.
[7] The Times of India. (2014). Sending vulgar messages, photographs
amount to outraging modesty: HC - Times of India. [online] Available
at: https://timesofindia.indiatimes.com/india/Sending-vulgar-messages-
photographs-amount-to-outraging-modesty-
HC/articleshow/28286434.cms.
[8] DNA. (2014). Engineer arrested for sending obscene picture to woman |
Latest News & Updates at Daily News & Analysis. [online] Available
at: http://www.dnaindia.com/mumbai/report-engineer-arrested-for-
sending-obscene-picture-to-woman-1983464.
Fig. 4. Encrypted data stored in Firebase Server
[9] S. Karthick and S. Binu, ‘Android Security Issues and Solutions’,
International Conference on Innovative Mechanisms for Industry
Applications (ICIMIA-IEEE), pp. 686-689, 2017.
V. CONCLUSION
[10] S. Karthick and S. Binu, ‘Static Analysis Tool for Identification of
It is the need of the hour to create a professional application Permission Misuse by Android Applications’, International Journal of
that will not let the user send inappropriate messages. The Applied Engineering Research, 12(24), pp.15169-178, 2017
proposed application developed in one of the most widely [11] Er. Kavindra Singh, Enhanced Education Chat Application Based on
Interested Keyword with Username and Password Authentication
used operating systems in the world that is an Android OS and Security. (2017). International Journal of Advanced Research in
cryptographic techniques incorporated. Computer Science and Software Engineering, 6(6).
[12] Nikita Mahajan, Design of Chatting Application Based on Android
Bluetooth. (2017). International Journal of Computer Science and
Mobile Computing, 3(3).
ACKNOWLEDGMENTS [13] Satrya, G., Daely, P. and Shin, S. (2016). Android forensics analysis:
Private chat on social messenger. 2016 Eighth International Conference
This research is conducted at Christ University, Bengaluru, on Ubiquitous and Future Networks (ICUFN).
Karnataka, India. We would like to thank Dr. Surendra [14] Shi, J., Mowery, D., Zhang, M., Sanders, J., Chapman, W. and Gawron,
Umesh Kulkarni, Dean of Sciences for his support for the L. (2017). Extracting Intrauterine Device Usage from Clinical Texts
research work. We would also like to thank Prof. Joy Using Natural Language Processing. 2017 IEEE International
Paulose, HOD, Department of Computer Science for his Conference on Healthcare Informatics (ICHI).
consistent guidance, support and encouragement throughout [15] Di Lecce, V., Calabrese, M., Soldo, D. and Quarto, A. (2010). Dialogue-
the research work. We would also like to thank Dr. Rohini V, oriented interface for linguistic human-computer interaction: A chat-
based application. 2010 IEEE International Conference on Virtual
Coordinator of MCA for her support for the research work. Environments, Human-Computer Interfaces and Measurement Systems.
[16] Forsythand, E. and Martell, C. (2007). Lexical and Discourse Analysis
of Online Chat Dialog. International Conference on Semantic
REFERENCES Computing (ICSC 2007).
[17] Firebase. (2017). User Based Security | Firebase Realtime Database |
[1] Most popular mobile messaging apps worldwide as of January 2017, b. Firebase. [online] Available at:
(2017). Most popular messaging apps 2017 | Statista. [online] Statista. https://firebase.google.com/docs/database/security/user-security.
Available at: https://www.statista.com/statistics/258749/most-popular-
global-mobile-messenger-apps/. [18] Firebase. (2017). Learn to Secure Files | Firebase. [online] Available at:
https://firebase.google.com/docs/storage/security/secure-files.
[2] Internetdemocracy.in. (2017). Section 66A: Do not send offensive
messages – The Internet Democracy Project. [online] Available at: [19] SearchSecurity. (2017). What is MD5? - Definition from WhatIs.com.
https://internetdemocracy.in/laws/the-information-technology- [online] Available at:
amendment-act-2008/section-66a/. http://searchsecurity.techtarget.com/definition/MD5.
[3] Avasarala, S. (2015). AN ANALYSIS OF VALIDITY OF SECTION
66A OF IT ACT, 2000 IN SHREYA SINGHAL V. UNION OF INDIA.