See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/360801505

Design of Chatbot System for College Website

Article in INTERNATIONAL JOURNAL OF COMPUTER SCIENCES AND ENGINEERING · July 2021

DOI: 10.26438/ijcse/v9i7.4145

CITATIONS READS

0 89

5 authors, including:

Pushkar Shinde Sopan Talekar

NDMVPS's KBT COE , Nasik Nashik District Maratha Vidya Prasarak Samaj's K.B.T. College of Engineering
7 PUBLICATIONS 3 CITATIONS 15 PUBLICATIONS 10 CITATIONS

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Medical Data Mining View project

Cognitive Radio Network View project

All content following this page was uploaded by Sopan Talekar on 13 January 2023.

The user has requested enhancement of the downloaded file.

 International Journal of Computer Sciences and Engineering Open Access
Research Paper Vol. 9, Issue.7, July 2021 E-ISSN: 2347-2693

Design of Chatbot System for College Website

Shivani Pravin Rashinkar1*, Neha Dilip Wanjol2, Shivani Nandkumar Rane3, Pushkar P.
Shinde4, Sopan A. Talekar5
1,2,3,4,5
Department of Computer Engineering, MVP’s KBT College of Engineering, Pune University, Nashik, India

*Corresponding Author: [email protected], Phone no.:+918805871670

DOI: https://doi.org/10.26438/ijcse/v9i7.4145 | Available online at: www.ijcseonline.org

Received: 14/Jul/2021, Accepted: 20/Jul/2021, Published: 31/Jul/2021

Abstract— Most of the time, Students need to visit college administration office to collect various information regarding
college such as Tuition fees, Term Schedule, etc. during admission process or as per their daily needs. Hence, to overcome
this problem, a chatbot can be designed and developed which can be easily integrated with any college website to provide
necessary information regarding college. The goal of AI based chatbot is to make an efficient conversation between human
and machine via auditory or textual methods. This project uses Natural language processing to process the user’s query and
generate a meaningful response. Based on the information stored in the database, bot itself determines appropriate response
of a particular query fired by user. The Chatbot is based on an Artificial Intelligence algorithm, which analyses user’s
question and responds with a Naive Bayes’ algorithm. This system will be a Web Application and can reduces work of
college administration providing information to students. It also reduces the workload on the staff to answer all the queries
of the students.
Keywords— Chatbot, Query, Graphical User Interface, Natural Language Processing, Artificial Intelligence.

I. INTRODUCTION This paper is divided into several sections, where Section I

contains the introduction of chatbot system, Section II
Chatbot is a computer application that mimics human contains Related Work of chatbot system, Section III
conversations in its natural format along with textual or explains the Methodology with architecture diagram and
voice communication. In the utilization of AI techniques flow chart, Section IV contains Results and Section V
together with natural language processing (NLP) [1],[3] describes Conclusion and Future Work.
chatbot for college website can be designed. This system
will be a web application, so it can give solutions to the II. RELATED WORK
analysed queries of the user. User simply need to raise the
K. Bala, M. Kumar, S. Hulawale, and S. Pandit et al.
query to the chatbot and the system will reply to the users
[1] Project on Chatbot for college management is
through the powerful Graphical User Interface (GUI)
developed with the help of AI algorithms which can
which is similar to messaging application interface, and
analyse user’s queries. This is a web application that will
provides a friendly environment to the user as they are
give answers to the analysed queries of the user. Users will
much aware of operating messaging application[1],[2]. The
simply need to select the class for queries and ask the
user can submit the question about the college-related
question to the bot. In this paper, they have used Porter
information or activities such as admission process, contact
Stemmer algorithm to answer the user' queries. The Users
information, address, annual day, sports day, intake and
should register and login to the system. Once login, user
other cultural activities of colleges. Developing a chatbot
can access the various helping pages through which the
solves the problems that can arises in gathering required
user can ask queries related to college activities.
college information. This system can be accessible from
anywhere and anytime. Chatbot will deliver efficient and P. Nikhila, G. Jyothi, K. Mounika, Mr.K. K. Reddy et
relevant response to the user corresponding to their entered al [2] The chatbot named UNIBOT is designed for the
message. Chatbot system will be beneficial for students, students to ask university related question. This system
parents, teaching and non-teaching staff as well. Presently, uses the concept of Artificial Intelligence and Machine
there are various chatbots available for the students like Learning. The System uses PHP Language for the
UNIBOT, ALICE etc. UNIBOT is designed for the development of UNIBOT. The query is given as an input
students to ask university related question. For this system to the algorithm, which processes it and gives the
a new algorithm is developed to deliver an appropriate corresponding response to the user. The GUI is similar to a
response to the user corresponding to their entered Messaging Application. It delivers efficient and relevant
message[2]. ALICE is a rule-based chatbot based on the response to the user corresponding to their entered query.
Artificial Intelligence Markup Language (AIML). This New algorithm is developed for UNIBOT. It is very
System uses NLP and Pattern Matching Algorithm efficient, requires less memory and has minimal database
[3],[6],[7] to process user’s query[8]. hits.

© 2021, IJCSE All Rights Reserved 41

 International Journal of Computer Sciences and Engineering Vol.9(7), Jul 2021, E-ISSN: 2347-2693

B.Setiaji and F. W. Wibowo et al. [3] Chatterbot is by the user. For the chatbot development, we have used
designed with a powerful pattern matching algorithm. This Python programming language using Django framework
project uses Indonesian conversational pattern and MySQL and Chatterbot library. It makes it easy to generate
database. This application is based on a knowledge base automated responses to a user’s input with the help of a
which is maintained by admin. It can be miss in defining a machine learning algorithm to produce different types of
sentence and how to response it while connecting chat responses. Students just have to query through the bot
application to the database. In the pattern-matching which is used for chatting purpose. Chatbot will reply to
operation, knowledge representation and implementation the query with the help of artificial intelligence.
of SQL are important. A data that has been created which
is based on the pattern of the conversation must be tested The proposed system will reduce the administration burden
by the help of a series of scenarios. The conversation and will be able to provide necessary details to students
should be crosschecked to the basic pattern so that it and parents online. Students do not have to visit college
allows you to add some knowledge to the database which administration for every inquiry. Students will get their
is not added before. If the input record in the database queries resolved without any hassle to reach out the college
does not match, then it will be remodeled. administration office. The System will be available for
24/7 to all students and parents [4].
K. Shivam, K. Saud, M. Sharma, S. Vashishth, and S.
Patil et al. [4] In this paper, for designing chatbot, 1. User Login:
Facebook Messenger is used which is source and uses User just needs to submit his/her queries regarding the
artificial intelligence to communicate with the user and college to the bot. User can be student, teacher as well as
provide the required information. This Facebook API is parents.
integrated with Python backend, webhook is used to
deliver the query of the user to the server. This system has 2. Chatbot Responding System:
used WIT and AI as a pre-trained artificial intelligence 2.1 NLP Processing for Query Analysis :
module so that one could use its pre-trained neural When a user submitted the query to the system, NLP is
networks to answer the user’s query with efficiency and applied and sense of the query is detected. Natural
accuracy. Language Processing (NLP) allows chatbot to understand
user’s messages and respond appropriately.
E. Haller and T. Rebedea et al. [5] This paper describes
the concept of identifying vital facts in texts describing the 2.2 Search Questions in database:
life of a historical figure for building a conversational Once the query is submitted, furthermore, we have to
agent that could be used in middle-school CSCL scenario. search the exact answer of the query in our pre-trained
This paper presents a way for building a chat-bot that may database. As the query description can change from person
simulate a historical figure. The system can receive "input" to person. The same question can be asked in different
as an understandable text or a web page about the historical ways from multiple users. One user asks a question so
figure and has as "output" a trained conversational agent simply and clearly, while another user may ask the same
that is in a position to answer all reasonably questions question with different format. So it is necessary to find
about the life expertise of that user. out what is the correct solution of a submitted query.

S. B. Sonawane, A. S.Badwar, R. H. Dalvi, G. N. More 2.3 Answer the Queries:

and S. A. Talekar et al. [6] This paper describes the As described above, whenever user submits a complaint,
concept of AI based chatbot which is designed for Student then it is checked that is there such question registered in
Counselling for career guidance. This system uses NLP database. If the answer is matched, then that is sent to that
and Keyword Matching Algorithm to process user’s query. User. If the answer of submitted query is not found in the
This System uses a modular architecture to respond to user database, then such questions are answered by admin
input. Each module contains knowledge based person. Once he answered the query, the answer is sent to
initialization mechanism, and logic to handle user requests. that user. And also it is stored in database so that whenever
The students are able to get proper guidance for career in such questions will be asked they get answer of that query
the field of their choice, also the college list for the same is directly from the database. So admin doesn’t need to
provided as per requirement. answer the same query again and again. For this we are
using powerful Naive Bayes’ Algorithm[9].
III. METHODOLOGY
3. Architecture Diagram of Chabot System:
The proposed chatbot system is a web application which Figure 1: shows the architecture of the chatbot system for
gives reply to the question of the user. This system is college website.
utilized for talking. A chatbot project is built using
artificial algorithms i.e. Naive Bayes’ algorithm that
analyse user requests and understand the user's message[9].
The system uses Natural Language Processing (NLP) and
built-in artificial intelligence to answer the queries asked

© 2021, IJCSE All Rights Reserved 42

 International Journal of Computer Sciences and Engineering Vol.9(7), Jul 2021, E-ISSN: 2347-2693

5. Algorithm Implemented
Naive Bayes’ is a powerful algorithm for text classification
problems. It is a probabilistic machine learning algorithm
which is based on Bayes’ theorem[9]. This classifier
assumes that the presence of a selected function in a
category is unrelated to the presence of another function. In
this algorithm a closed domain dataset containing
questions/user-responses and corresponding answers is
made, which every question is given a label, this will relate
the question to its answer. Due to multiple questions could
have the same response, there can be multiple questions
having the same answer.

The Formula for Naive Bayes’ Algorithm is as follows:

P (A | B) = (P (B | A) * P(A))

P(B)

Where,
Figure 1: Architecture Diagram of Chatbot System
P (A | B) = Probability of ‘A’ occurring given
evidence of ‘B’ has already occurred.
4. Flowchart of Chatbot System:
The flowchart of the system displays how the chatbot P (B | A) = Probability of ‘B’ occurring given
performs. Initially, the user message is pre-processed and evidence of ‘A’ has already occurred.
connectivity to the database is obtained. Then, based upon
conditions satisfied, the chatbot process flows and provide P (A) = Probability of ‘A’ occurring.
response to the user. If the user cannot find the answer for
a query then in such condition chatbot will provide P (B) = Probability of ‘B’ occurring.
admin’s contact details to the user. Admin can view and
answer the corresponding query. 6. Different Algorithms Used for Chatbot System:
Some of the most popular algorithms for chatbots are
Porter Stemmer[1], Naive Bayes’[9], Support Vector
Machines, K-means and atural language processing (NLP).
Chatbots mainly use classification algorithms to recognize
intent in phrases. Every algorithm has its own advantages
and disadvantages according to its working method. Table
1 shows comparison of different algorithms WRT to Naive
Bayes’ algorithm.

Table 1: Comparison of Different Algorithms

Sr. No. Porter- K-Means Naive Bayes’

Stemmer Clustering Algorithm
Algorithm Algorithm

1. The stems Different Naive Bayes’

generated partitions can classifier performs
are not result into better than other
always real different final models with less
words. clusters. training data if the
assumption of
independence of
features holds.
2. It has five Difficult to The algorithm
steps and predict K- works very fast
sixty rules. Value, hence and can easily
And takes it is time- predict the class of
more time. consuming. a test dataset.
Figure 2: Flowchart of Chatbot System

© 2021, IJCSE All Rights Reserved 43

 International Journal of Computer Sciences and Engineering Vol.9(7), Jul 2021, E-ISSN: 2347-2693

3. This System Doesn’t The algorithm V. CONCLUSION AND FUTURE SCOPE

is limited to support performs well with
English categorical categorical The aim of the system is to provide a user-friendly and
words only. data. variables in efficient chatbot system for College. The chatbot will be
comparison to
very useful in guiding students to get correct and up-to-
numerical
variables. date information source. This system will be fruitful for
students, teachers as well as parents. They can get
IV. RESULTS AND DISCUSSION information at any time without having to visit the college
administration office every time. In the future scope of this
The Chatbot is to carry out a conversation between both project, the system will include voice-based requests and
human and machine. Figure 3: shows the simple and responses. Users are required to provide voice input and
attractive pop-up Interface for College Chatbot System. the system will output in the form of text. Chatbot will be
Firstly, chatbot will print a welcome message then a user able to provide a voice output as well, with the help of text
can type and submit the query and bot will provide an to speech or speech to text conversion.
appropriate answer to the user’s query. In figure 4 ‘What is
college full name?’ this question is asked by the user and REFERENCES
correct answer is given by the chatbot. The User can query
[1] K. Bala ,M. Kumar, S.Hulawale, and S. Pandita, “Chat-Bot For
any college related activities through the chatbot system. College Management System Using A.I”, International Research
Journal of Engineering and Technology (IRJET), Vol. 04,
Issue.11, pp.2395-0072, 2017.
[2] P. Nikhila, G. Jyothi, K. Mounika, Mr.K. K. Reddy and Dr.
B.V. Ramana Murthy, “AI and Web-Based Human-Like
Interactive University Chatbot (UNIBOT)”,In theProceedings of
the Third International Conference on Electronics
Communication and Aerospace Technology[ICECA 2019],
pp.1-12,2019.
[3] B.Setiaji, F. W. Wibowo, "Chatbot Using A Knowledge in
Database- Human-to-Machine Conversation Modeling", In the
Proceedings of the 2016 International Conference on Intelligent
Systems, Modelling and Simulation, pp.2166-0670, 2016.
[4] K. Shivam, K. Saud, M. Sharma, S. Vashishth, and S. Patil,
“Chatbot for College Website”, International Journal of
Computing and Technology, Vol.5, Issue.6, pp.2348-6090,
2018.
[5] E. Haller and T. Rebedea, "Designing a Chat-bot that Simulates
a Historical Figure", Faculty of Automatic Control and
Computers university of Bucharest, IEEE 978-0-7695-4980-
4/13, 2013.
[6] S. B. Sonawane, A. S.Badwar, R. H. Dalvi, G. N. More and S.
A. Talekar, “Design of Chatbot System for Student
Counselling”, International Journal of New Innovations in
Engineering and Technology, Vol.13, Issue.3, pp.2319-6319,
Figure 3: Chatbot Interface-1
2020.
[7] P. Jain,“College Enquiry Chatbot Using Iterative Model",
International Journal of Scientific Engineering and Research
(IJSER),Vol.7, Issue.1, pp.2347-3878,2019.
[8] N. Hatwar, A. Patil and D. Gondane, “AI BASED CHATBOT”,
International Journal of Emerging Trends in Engineering and
Basic Sciences (IJEEBS) ISSN (Online) 2349-6967 Volume 3,
Issue 2 (March-April 2016).
[9] A. Prajapati, P. Naik, S. Singh and A. Kale, “Android Based
Chatbot For College”,International Journal of Scientific &
Engineering Research, Vol.9, Issue.4, pp.2229-5518, 2018.
[10] A. Ohm, K. Bhavani,“Chatbot for Career Guidance Using AI”,
International Journal of Computer Sciences and Engineering
(IJCSE), Vol.7, Issue.6 ,pp.2347-2693, 2019.
[11] K. G. Yohandi, M. I. Sani, “Designing a Mobile Chatbot For
Elementary School Vocabulary”, International Journal of
Computer Sciences and Engineering(IJCSE), Vol.8, Issue.7,
pp.2347-2693, 2020.

Figure 4: Chatbot Interface-2

© 2021, IJCSE All Rights Reserved 44

 International Journal of Computer Sciences and Engineering Vol.9(7), Jul 2021, E-ISSN: 2347-2693

AUTHORS PROFILE
Shivani Pravin Rashinkar, Bachelor
of Engineering in Computer Science,
MVP’s KBT College of Engineering,
Nashik, India

Neha Dilip Wanjol, Bachelor of

Engineering in Computer Science,
MVP’s KBT College of Engineering,
Nashik, India

Shivani Nandkumar Rane, Bachelor

of Engineering in Computer Science,
MVP’s KBT College of Engineering,
Nashik, India

Mr. Pushkar P. Shinde, pursed

Bachelor of Engineering in Computer
Science and Master of Engineering in
Computer Science. Completed
Diploma in Advance Computing,
CDAC. He is currently working as
Assistant Professor in Department of
Computer Engineering, MVPS’s
KBTCOE, Nashik, India. He has published & presented
more than 06 research papers. His research interest is
Machine Learning. He has 16 years of teaching experience.

Mr. Sopan A. Talekar, pursued

Bachelor of Engineering from North
Maharashtra University Jalgaon,
Maharashtra in 2004 and Master of
Technology from Visvesvaraya
Technological University, Belagavi,
Karnataka in year 2010. He is
currently pursuing Ph.D and currently
working as Assistant Professor in Department of Computer
Engineering, MVPS’s KBTCOE, SPPU, Pune, since 2005.
He is a life member of IE & ISTE. He has published &
presented more than 07 research papers. He received
prestigious Promising Engineer Award from the Institution
of Engineers (India), Local center Nasik, in 2013. His
research interest is Wireless Network, Mobile Ad hoc
Network, Cognitive Radio Network & Artificial
Intelligence. He has 16 years of teaching experience.

© 2021, IJCSE All Rights Reserved 45

View publication stats