COLLEGE ENQUIRY CHATBOT

A DESIGN PROJECT REPORT

Submitted by

SARASWATHY P L

SHANMATHI S

UMA MAHESWARI K

VEDHA VARSHINI V

in partial fulfillment for the award of the degree of

BACHELOR OF ENGINEERING
in

COMPUTER SCIENCE AND ENGINEERING

K. RAMAKRISHNAN COLLEGE OF TECHNOLOGY

(An Autonomous Institution, affiliated to Anna University Chennai and Approved by AICTE, New Delhi)

SAMAYAPURAM – 621 112

NOVEMBER, 2024
 COLLEGE ENQUIRY CHATBOT

A DESIGN PROJECT REPORT

Submitted by

SARASWATHY P L (811722104135)

SHANMATHI S (811722104141)

UMA MAHESWARI K (811722104170)

VEDHA VARSHINI V (811722104178)

in partial fulfillment for the award of the degree

of

BACHELOR OF ENGINEERING
in

COMPUTER SCIENCE AND ENGINEERING

K. RAMAKRISHNAN COLLEGE OF TECHNOLOGY

(An Autonomous Institution, affiliated to Anna University Chennai and Approved by AICTE, New Delhi)

SAMAYAPURAM – 621 112

NOVEMBER, 2024

i
 K RAMAKRISHNAN COLLEGE OF TECHNOLOGY
(AUTONOMOUS)
SAMAYAPURAM – 621 112

BONAFIDE CERTIFICATE

Certified that this project report titled “COLLEGE ENQUIRY CHATBOT” is

bonafide work of the students SARASWATHY P L(811722104135), SHANMATHI
S(811722104141), UMA MAHESWARI K(811722104170), VEDHA VARSHINI
V(811722104178) who carried out the project under my supervision. Certified further,
that to the best of my knowledge the work reported here in does not form part of any
other project report or dissertation on the basis of which a degree or award was conferred
on an earlier occasion on this or any other candidate.

SIGNATURE SIGNATURE
Dr.A.Delphin Carolina Rani M.E.,Ph.D., Mr. M Saravanan M.E.,
HEAD OF THE DEPARTMENT SUPERVISOR
PROFESSOR Assistant Professor Department of CSE
Department of CSE K Ramakrishnan College of Technology
K Ramakrishnan College of Technology (Autonomous)
(Autonomous) Samayapuram – 621 112
Samayapuram – 621 112

Submitted for the viva-voice examination held on ………………

INTERNAL EXAMINER EXTERNAL EXAMINER

ii
 DECLARATION

We jointly declare that the project report on “COLLEGE ENQUIRY CHATBOT” is

the result of original work done by us and best of our knowledge, similar work has not
been submitted to “ANNA UNIVERSITY CHENNAI” for the requirement of Degree
of Bachelor Of Engineering. This project report is submitted on the partial fulfilment of
the requirement of the award of Degree of Bachelor Of Engineering.

Signature

SARASWATHY P L

SHANMATHI S

UMA MAHESWARI K

VEDHA VARSHINI V

Place: Samayapuram

Date:

iii
 ACKNOWLEDGEMENT

It is with great pride that we express our gratitude and indebtness to our
institution “K RAMAKRISHNAN COLLEGE OF TECHNOLOGY
(AUTONOMOUS)”, for providing us with the opportunity to do this project.

We are glad to credit honorable chairman Dr. K RAMAKRISHNAN, B.E., for

having provided for the facilities during the course of our study in college.

We would like to express our sincere thanks to our beloved Executive Director
Dr. S KUPPUSAMY, MBA, Ph.D., for forwarding our project and offering adequate
duration to complete it.

We would like to thank Dr. N VASUDEVAN, M. Tech., Ph.D., Principal, who

gave opportunity to frame the project with full satisfaction.

We whole heartily thank Dr. A DELPHIN CAROLINA RANI, M.E., Ph.D.,

Head of the Department, COMPUTER SCIENCE AND ENGINEERING for
providing her support to pursue this project.

We express our deep and sincere gratitude and thanks to our project supervisor
Mr. M. SARAVANAN, M.E., Department of COMPUTER SCIENCE AND
ENGINEERING, for his incalculable suggestions, creativity, assistance and patience
which motivated us to carry out this project.

We render our sincere thanks to Course Coordinator and other staff members for
providing valuable information during the course. We wish to express our special thanks
to the officials and Lab Technicians of our departments who rendered their help during
the period of the work progress.

iv
 ABSTRACT

A chatbot, usually referred to as a chatterbot, attempts to have a conversation with

a person. When a question is posed, the system has the ability to detect sentences and

select the proper answer. The response principle is the matching of the user's input

phrase. The current technical project involves building a professional system for a

college help desk employing an android-based chatbot, artificial intelligence

technology, and virtual assistance (human- machine communication), then sending that

natural language to a server. Chatbot systems have become increasingly popular for

automating interactions with users and providing information in various domains,

including college enquiries. In this paper, we propose a chatbot system for college

enquiry using a knowledgeable database. The system utilizes a knowledgeable database

that contains relevant information about the college, such as courses, faculty, campus

facilities, and admissions procedures. The system employs various algorithms,

including rule-based, retrieval-based, natural language processing (NLP), and machine

learning algorithms, to understand and respond to user queries in a context-aware

manner. The rule- based algorithms provide predefined rules and patterns for handling

specific intents or frequently asked questions, while the retrieval-based algorithms

search the knowledgeable database for relevant information.

v
 TABLE OF CONTENTS

Chapter No. TITLE Page No.

ABSTRACT V

LIST OF FIGURES viii

LIST OF TABLES ix

LIST OF ABBREVIATIONS x

1 INTRODUCTION 1

1.1 General Information 1

1.2 Problem statement 2

1.3 Objectives 3

1.4 System Architecture 4

1.5 Statement Scope 5

1.6 Natural Language Processing 6

2 LITERATURE SURVEY 9

2.1 Open problems in existing system 10

2.2 Inferences from literature survey 12

3 REQUIREMENT ANALYSIS 14

3.1 Software and Hardware Requirements Specification Document 14

3.2 System Use case 15

4 DESCRIPTION OF PROPOSED SYSTEM 16

4.1 Study of the Project 16

4.2 Existing Methodology 18

4.3 Proposed Methodology 23

4.4 Project Task Set/Project Management Plan 24

vi
5 IMPLEMENTATION DETAILS 25

5.1 Development and Deployment Setup 25

5.2 Algorithms 29

5.3 Module Implementation 35

5.4 Data Flow Diagrams 37

5.5 Use Case Diagram 39

5.6 Class Diagram 40

5.7 Sequence Diagram 41

5.8 Component Diagram 42

5.9 Deployment Diagram 43

5.10 Collaboration Diagram 44

5.11 State Chart Diagram 45

6 RESULTS AND DISCUSSION 46

7 CONCLUSION 50

7.1 Conclusion 50

7.2 Future work 51

APPENDIX 52

A. SOURCE CODE 52

B. SCREENSHOTS 59

REFERENCES 60

vii
 LIST OF FIGURES

FIGURE Page No.

FIGURE NAME
NO

1.1 System Architecture 4

4.1 Knowledge Canvas Diagram 21

5.1 Structure of LSTM 30

5.2 Forget Gate 31

5.3 Input Gate 31

5.4 Output Gate 32

5.5 Data Flow Diagrams 37

5.5.1 Level 0 Data Flow Diagram 37

5.5.2 Level 1 Data Flow Diagram 38

5.6 Use Case Diagram 39

5.7 Class Diagram 40

5.8 Sequence Diagram 41

5.9 Component Diagram 42

5.10 Deployment Diagram 43

5.11 Collaboration Diagram 44

5.12 State Chart Diagram 45

viii
 LIST OF TABLES

TABLE NO TABLE NAME Page No.

4.1 IDEA 18

4.2 IDEA Matrix 19

ix
 LIST OF ABBREVIATIONS

S.No ABBREVIATIONS EXPANSION

1 AI Artificial Intelligence
2 CNN Convolution Neural Network
3 GUI Graphical User Interface
4 LSTM Long Short Term Memory
5 NER Named Entity Recognition
6 NLG Natural Language Generation
7 NLP Natural Language Processing
8 NLTK Natural Language Tool Kit
9 UML Unified Modeling Language

x
 CHAPTER – 1

INTRODUCTION

1.1 GENERAL INFORMATION:

This Application is for college students, staff, and parents. Easy way to interaction and time
consuming. This project is mainly targeted at colleges and the synchronization of all the sparse and
diverse information regarding regular college schedule. Generally, students face problems in getting
correct notifications at the correct time, sometimes important notices such as campus interview,
training and placement events, holidays, and special announcements. Smart Campus tries to bridge
this gap between students, teachers, and college administrators. Therefore in the real world scenario,
such as college campus, the information in the form of notices, oral communication, can be directly
communicated through the android devices and can be made available for the students, teachers
directly for their android devices and the maintenance of application will be easier in later future
because of the use of architectural MVC which separates the major works in the development of an
application such as data management, mobile user interface display and web service which will be the
controller to make sure for fast and efficient maintenance of application.

The College bot project is built using artificial algorithms that analyses user’s queries and
understand user’s message. This System is a web application which provides answer to the query of
the student. Students just must query through the bot which is used for chatting. Students can chat
using any format there is no specific format the user has to follow. The System uses built in artificial
intelligence to answer the query. The answers are appropriate what the user queries. The User can
query any college related activities through the system. The user does not have to personally go to the
college for enquiry. The System analyses the question and then answers to the user.

The system answers to the query as if it is answered by the person. With the help of artificial
intelligence, the system answers the query asked by the students.The system replies using an effective
Graphical user interface which implies that as if a real person is talking to the user. The user just
must register himself to the system and has to login to the system.

1
 Based on the recent epidemiological situation, the increasing demand and relianceon electronic
education has become very difficult to access to the university due to the curfew imposed, and this has
led to limited access to information for academics at the university.

This project aims to build a chatbot for Admission and Registrationto answer every person who
asks about the university, colleges, majors, and admission policy. Artificial intelligence (AI) is a
branch of computer science that focuses on creating machines that can perform tasks that typically
require human intelligence, such as perception, reasoning, learning, and decision-making.

Natural Language Processing: AI algorithms enable chatbots to understand natural language inputs
from users and interpret them accurately.

Machine Learning: AI algorithms enable chatbots to learn from user interactions and improve their
responses over time. Machine learning algorithms analyze the data collected from user interactions
and identify patterns and trends. Based on this analysis, the chatbot can be trained to provide more
accurate and relevant responses.

Personalization: AI algorithms enable chatbots to personalize their responses based on user

preferences and behavior. By analyzing user data, chatbots can tailortheir responses to each user's
specific needs and preferences.

1.2 PROBLEM STATEMENT

At the start of each academic semester, registration opens for those wishing to jointhe university
in various disciplines, and telephone calls for admission and registration abound. This leads to an
increase in the loads and work for the employees of the Deanship of Admission and Registration as a
result of the constantpressure of those wishing to register and their families by flocking to the
Deanship, so the employees are not able to answer the phone calls and social media. This often leads
to many students who wish to register to be ignored.

The process of providing information and support to prospective and current students in a timely
and efficient manner is a challenge for colleges, leading to frustration and dissatisfaction among users.

2
 Knowledge database that contains all relevant information about the college and itsoperations. This
database should be regularly updated to ensure that the information provided by the chatbot is accurate.

1.3 OBJECTIVES

• Save effort and time for both the admission and registration staff and students who wish
to enroll.

• Provide detailed information about colleges and majors.

• Easy access to information.

• To minimize the time required to solve the queries.

• To give response to the user based on queries.

• To simplify communication between user and machine.

3
 1.4 SYSTEM ARCHITECTURE

Figure 1.1 System Architecture

A block diagram is a diagram of a system in which the principal parts or functions are
represented by blocks connected by lines that show the relationships of the blocks. It may also show
how the system operates, what are its inputs and outputs at various stages, and how the information,
and/or materials flow through it. The block diagram for” Online chatting system for college enquiry
knowledgeable Database” The proposed system has a client server architecture. All the information
will be kept in an optimized database on the central server. This information can be accessed by the
users through the android application installed on their smartphones (client machines). Each client
machine will have an improved user interface.

4
Modules Client-Server (chat user): The proposed system has a client server architecture. All the
information will be kept in an optimized database on the central server. This information can be
accessed by the users through the android application installed on their smartphones (client machines).
Each client machine will have an improved user interface.

Chatbot: A chatbot is a technology that allows users to have natural conversations to access content
and services. Chatbots typically take the form of a chat client, leveraging natural language processing
to conduct a conversation with the user. Chatbots control conversation flow based on the context of
theusers requests and respond with natural language phrases to provide direct answers, request
additional information or recommend actions that can be taken.

Pattern matching: Bot send a query to a machine for comparing. The query match with database
sends to data services.

Data Services: Intent is used to call upon proper service. using entity information to find proper data.
Hence all the modules described above are completed in polynomial time sec t, So the problem is P.

1.5 STATEMENT SCOPE

In today’s world as there are everything is digital. In education system work is very lengthy and
time consuming and required extra manpower. We develop this application for students, teachers,
parents, and guest. In this project we implement android application due to this application The Student
does not have to go personally to college office for the enquiry.

The application enables the students to be updated with college cultural activities. If application
saves time for the student as well as teaching and non-teaching staffs. It is useful for parents also to
show his/her child marks and important notices.

5
1.6 NATURAL LANGUAGE PROCESSING

NLP is an interdisciplinary subfield of linguistics, computer science, and artificial intelligence

concerned with the interactions between computers and human language, how to program computers
to process and analyze large amounts of natural language data. The goal is a computer capable of
"understanding" the contents of documents, including the contextual nuances of the language within
them. The technology can then accurately extract information and insights contained in the documents
as well as categorize and organize the documents themselves. NLP techniques allow the chatbot to
understand the natural language queries of users and provide accurate and relevant responses. NLP is
a critical component of many applications that involve language, such as chatbots, voice assistants,
machine translation, sentiment analysis, and more. It involves several techniques and approaches,
including statistical modeling, machine learning, deep learning, and rule- based systems, to analyze
and process natural language data. NLP is a rapidly evolving field, and recent advances in machine
learning and deep learning have led to significant improvements in the accuracy and performance of
NLP models. Challenges in natural language processing frequently involve speech recognition,
natural-language understanding, and natural-language generation.

In early days, many language-processing systems were designed by symbolic methods, i.e., the
hand-coding of a set of rules, coupled with a dictionary lookup such as by writing grammars or devising
heuristic rules for stemming.

More recent systems based on machine-learning algorithms have many advantages over
hand-produced rules:

• The learning procedures used during machine learning automatically focus on the most
common cases, whereas when writing rules by hand it is often not at all obvious where the effort
should be directed.

6
 • Automatic learning procedures can make use of statistical inference algorithms to produce
models that are robust to unfamiliar input (e.g. containing words or structures that have not
been seen before) and to erroneous input (e.g. with misspelled words or words accidentally
omitted). Generally, handling such input gracefully with handwritten rules, or, more generally,
creating systems of handwritten rules that make soft decisions, is extremely difficult, error-
prone and time- consuming.

• Systems based on automatically learning the rules can be made more accurate simply by
supplying more input data. However, systems based on handwritten rules can only be made
more accurate by increasing the complexity of the rules, which is a much more difficult task.
In particular, there is a limit to the complexity of systems based on handwritten rules, beyond
which the systems become more and more unmanageable. However, creating more data to
input to machine- learning systems simply require a corresponding increase in the number of
man-hours worked, generally without significant increases in the complexity of the annotation
process.

Despite the popularity of machine learning in NLP research, symbolic methods are still (2020)
commonly used:

• When the amount of training data is insufficient to successfully apply machine learning
methods, e.g., for the machine translation of low- resource languages such as provided by the
Apterium system,

• For pre-processing in NLP pipelines, e.g., tokenization, or

• For postprocessing and transforming the output of NLP pipelines, e.g., for knowledge
extraction from syntactic parses.

7
Sentiment analysis: Sentiment analysis is the process of identifying the emotion or sentiment behind
a user's query. This can be useful in chatbots designed to provide emotional support or mental health
services. Sentiment analysis is typically done using machine learning algorithms that are trained on
large datasets of text labeled with sentiment.

Text generation: Text generation techniques can be used to generate natural language responses to
user queries. These techniques use machine learning algorithms to analyze the context of the user's
query and generate a relevant response. Text generation can be particularly useful in chatbots designed
to handle complex or multi-turn conversations.
Natural Language Processing (NLP) techniques can play a crucial role in developing a college enquiry
chatbot.

Intent recognition: NLP techniques can be used to recognize the intent of user queries, allowing the
chatbot to provide appropriate responses.

Named entity recognition (NER): NER is a subtask of entity recognition that focuses specifically on
identifying named entities such as people, organizations, andlocations. NER is commonly used in
chatbots designed for customer service or support, where identifying customer names or order numbers
is important.

Language translation: Chatbots can be designed to provide multilingual support by using language
translation techniques. Machine learning algorithms are trained on large datasets of text in multiple
languages to provide accurate translations.

Overall, NLP techniques are critical to the success of chatbots, allowing them to accurately understand
user queries and provide relevant and personalized responses.

8
 CHAPTER – 2

LITERATURE SURVEY

Professor Girish Wadhwa suggested that the institution build an inquiry chatbot using artificial
intelligence in March-April 2017. Algorithms that might analyze consumer inquiries and recognize
consumer messages. This machine might be a chatbot with the intention to provide solutions to
students' questions. Students actually need to pick out a category for department requests and then
request a botto be used for chat. The project's main goal is to develop an algorithm that may be used to
correct the answers to queries that customers ask. It is essential to create a database where all related
statistics can be kept as well as to expand the online interface. A database can develop to be able to
compile information on queries, responses, key words, logs, and messages. 2016 saw Bayu Setiaji
publish "Chatbot the usage of database knowledge." A chatbot is made to communicate with
technology.

Machine learning is built to recognize sentences and concluded, such as the answer to a
question. Personalized message, i.e. A request is saved in accordance with theresponse. The more
similarly the statements are stated, the more it will be marked as similarity of the sentences. It is then
answered in light of the answers from the first sentence. The sentence similarity calculator breaks the
input sentence down into its component letters. A database stores the knowledge of chatbots. A chatbot
has interfaces, and the database control system's access point through this interface is at its core. The
Chatbot application was created using a variety of programming languages with the addition of a user
interface that allows users to give input and get a response. Starting with the symbol of entity date,
which produced 11 entities and their cardinalities, the structure and building of tables was done as an
indication of the knowledge contained inside the database. SQL was used in a way that was tailored to
the model that was kept inside the programme.

Scripts was presented by Md. Shahriare Satu and Shamim-Ai- Mamun. They asserted that
entirely AIML-based chatbots are easy to set up, lightweight, and eco- friendly to use. post provides
information on the various ways that chatbots are used. An AIML and LSA based chatbot was created
by ThomasN. T. and Amrita Vishwa to provide customer support on e-commerce platforms.

9
2.1 OPEN PROBLEMS IN EXISTING SYSTEM

There are several open problems that need to be addressed in college enquiry chatbots to
improve their performance and provide better user experience. Here aresome of the key open problems
in college enquiry chatbots:

Intent Identification: One of the primary challenges in developing a college enquirychatbot is

accurately identifying the user's intent. College enquiries can cover a wide range of topics, and the
chatbot needs to correctly identify the user's intention to provide an appropriate response.

Accuracy: While chatbots can provide quick and convenient access to information, they are not always
accurate in their responses. This is because the chatbot's database may not always be up-to-date, or the
natural language processing algorithms may not be able to correctly interpret the user's queries.

Knowledge Base Management: A college enquiry chatbot needs to have access to a

large amount of information about the college, including admission criteria, course offerings, faculty,
and campus facilities. Managing this knowledge base is a significant challenge, as the information is
often dispersed across multiple sources and needs to be kept up-to-date.

Natural Language Processing: Chatbots need to be able to understand and process natural language
inputs accurately. However, natural language processing (NLP) technology is still in its early stages,
and there are many challenges in accurately interpreting the meaning of user queries.

Chatbots can be integrated with a variety of channels, including websites, social media
platforms, and messaging apps, to provide convenient access to information. Natural language
processing (NLP) and machine learning (ML) techniques are commonly used in college enquiry
chatbots to understand user queries and providerelevant responses. The use of chatbots in college
enquiry systems can help to reduce workload for administrative staff and free up time for more
complex tasks.

10
 Personalization is an important feature of successful college enquiry chatbots, and can be
achieved through the use of user data and machine learning techniques. Chatbots can help colleges to
collect valuable data on user behaviour and preferences, which can be used to improve the quality of
the chatbot's responses.

Several studies have focused on designing chatbots that can simulate human- like conversations
and provide a personalized experience to users.

Multilingual Support: Colleges often have students from different parts of the world, speaking
different languages. Providing multilingual support in college enquiry chatbots is a challenge that
requires advanced NLP capabilities and a well-designed language model.

Personalization: To provide a better user experience, college enquiry chatbots need to personalize
their responses based on the user's profile, preferences, and history. This requires advanced machine
learning algorithms that can analyze userdata and provide tailored responses.

Context Management: College enquiries often involve complex and multi- turn conversations.
Chatbots need to be able to maintain context across these conversations to provide accurate and
relevant responses.

User Engagement: Finally, chatbots need to be engaging and interactive to keep users interested and
motivated to continue using them. This requires designing chatbots that can simulate human-like
conversations and provide relevant and interesting information to users.

11
2.1 INFERENCES FROM LITERATURE SURVEY
Based on a literature survey of college enquiry chatbots, several key inferences canbe drawn:
College enquiry chatbots are becoming increasingly popular: There is a growing trend of
colleges and universities adopting chatbots to handle student enquiries. Several studies have shown
that chatbots can significantly reduce the workload on college administrators and provide faster,
more efficient.

Natural language processing (NLP) is a critical component of college enquiry chatbots: NLP
technology is used to understand and interpret user queries, and to generate natural language responses.
Several studies have focused on improving the accuracy and effectiveness of NLP in college enquiry
chatbots.

Machine learning (ML) algorithms are being used to improve the performance of college enquiry
chatbots: ML algorithms are used to train chatbots on large datasets of student queries and responses.
This helps chatbots to learn from past interactions and provide more accurate and relevant responses.

Chatbots are being used to support a wide range of college enquiries: College enquiry chatbots
can handle a wide range of enquiries, including admission inquiries, course registration, financial aid,
campus facilities, and career services. The success of college enquiry chatbots depends on effective
design and development, including careful consideration of user needs, the use of appropriate testing
and optimization.

Multilingual support is a growing area of research in college enquiry chatbots: Many colleges and
universities have a diverse student population, and providing multilingual support is essential to ensure
that all students can access the information and services they need.

12
Context management is a critical challenge in college enquiry chatbots: College enquiries often
involve complex and multi-turn conversations. Chatbots need to be able to maintain context across
these conversations to provide accurateand relevant responses.

User experience is a critical factor in the success of college enquiry chatbots: Chatbots need to be
engaging, interactive, and easy to use to keep students interested and motivated to use them. Several
studies have focused on designing chatbots that can simulate human-like conversations and provide a
personalized experience to users. College enquiry chatbots can help colleges to provide more efficient
and effective customer service to prospective and current students, and canimprove the overall user
experience.

Chatbots can be integrated with a variety of channels, including websites, social media
platforms, and messaging apps, to provide convenient access to information. Natural language
processing (NLP) and machine learning (ML) techniques are commonly used in college enquiry
chatbots to understand user queries and providerelevant responses. The use of chatbots in college
enquiry systems can help to reduce workload for administrative staff and free up time for more
complex tasks.

Personalization is an important feature of successful college enquiry chatbots, and can be

achieved through the use of user data and machine learning techniques. Chatbots can help colleges to
collect valuable data on user behaviour and preferences, which can be used to improve the quality of
the chatbot's responses.

13
 CHAPTER - 3

REQUIREMENT ANALYSIS

3.1 SOFTWARE AND HARDWARE REQUIREMENTS SPECIFICATION

DOCUMENT

SOFTWARE AND HARDWARE

REQUIREMENTS:

Hardware:

Operating system : Windows 7 or 7+

RAM : 2 GB MEMORY

Hard disc or SSD : More than 500 GB

Processor : Processor Dual Core

Software:

Software’s : Python 3.6 or high version

IDLE : PyCharm.

Framework : Flask

14
3.2 SYSTEM USE CASE

A college enquiry chatbot can have several use cases, including:

Admission Enquiry: The chatbot can provide information about the admissionprocess,
eligibility criteria, important dates, and documents required for admission.

Course Information: The chatbot can provide detailed information about the courses offered
by the college, including the duration of the course, syllabus, fees, and career opportunities.

Campus Facilities: The chatbot can provide information about the various facilitiesavailable
on the college campus, such as libraries, laboratories, sports facilities, andaccommodation options.

Fees and Scholarships: The chatbot can provide information about the fees structure for
different courses and scholarships available for students based on theiracademic performance.

Important Dates: The chatbot can remind students about important dates such asadmission
deadlines, fee payment dates, and exam schedules.

FAQs: The chatbot can answer frequently asked questions by students, such as how to apply
for admission, how to check the admission status.

Student life: The chatbot can provide information about student life at the college, including
clubs and societies, extracurricular activities, and student resources.

Counseling: The chatbot can provide counseling to students regarding their careeroptions,
course selection, and academic performance.

Academic support: The chatbot can assist students with academic enquiries, including course
registration, exam schedules, and study resources.

Admission and enrolment enquiries: The chatbot can assist prospective students with
admission and enrolment enquiries, including deadlines, application requirements, and documentation.
Overall, a college enquiry chatbot can provide aseamless and hassle-free experience for students who
are looking for information about the college and its courses.

15
 CHAPTER – 4

DESCRIPTION OF PROPOSED SYSTEM

4.1 STUDY OF THE PROJECT

This project is mainly targeted at colleges and the synchronization of all the sparseand diverse
information regarding regular college schedule. Generally, students face problems in getting correct
notifications at the correct time, sometimes important notices such as campus interview, training and
placement events, holidays and special announcements. Smart Campus tries to bridge this gap between
students, teachers, and college administrators. Therefore in the real worldscenario, such as college
campus, the information in the form of notices, oral communication, can be directly communicated
through the android devices and can be made available for the students, teachers directly for their
android devices and the maintenance of application will be easier in later future because of the use of
architectural MVC which separates the major works in the development of an application such as data
man agreement, mobile user interface display and web service which will be the controller to make
sure for fast and efficient maintenance of application.

A study is carried out to select the best system that meets the performance requirements. Feasibility is
the determination of whether a project is worth doing or not. The process followed in making this
determination is called a feasibility study. This type of study determines if a project can and should be
taken. Since the feasibility study may lead to the commitment of large resources, it becomes necessary
that it should be conducted competently and that no fundamental errors of judgment are made.
Depending on the results of the initial investigation, the survey is expanded to a more detailed
feasibility study. Feasibility study is a test of system proposal according to its work- ability, impact on
the organization, ability to meet user needs, and effective use of resources. The objective of the
feasibility study is not to solve the problem but to acquire a sense of its scope. During the study,
the problem definition is crystallized and aspects of the problem to be included in the system are
determined.

16
 Save timing of students and teachers and also save extra manpower. Student can see all
document related college like, notice, study material, question papers etc. on time to time and from
any place whether student is present in college or not. And also reduce the work of staff. It is proper
communication in between staff and students.

Natural language processing algorithms: To interpret user queries and generate accurate
responses.

Knowledgeable database: To store information about college programs, courses, and admission
requirements.

Recommendation engine: To suggest suitable programs based on the user's interests and
qualifications.

User interface: To provide a user-friendly and intuitive interface for users to interactwith the
chatbot.

Data collection and processing: To gather and organize information about college programs,
courses, and admission requirements.

Algorithm development: To develop natural language processing algorithms that can interpret
user queries and generate accurate responses.

Database design and implementation: To design and implement a knowledgeable database that
can store and retrieve information about college programs, courses, and admission requirements.

User interface design and implementation: To design and implement a user interface that is
intuitive and user-friendly.

Testing and evaluation: To test the chatbot system for accuracy, usability, and performance.

17
4.2 EXISTING METHODOLOGY

Knowledge graph creation: The first step is to create a knowledge graph that contains all the
relevant information about college programs, courses, and admission requirements. This can be done
using existing ontologies or by manuallycurating the knowledge graph.

4.2.1. To develop the problem under consideration and justify feasibility usingconcept of
knowledge canvas and IDEA matrix.

I D E A
Increase Drive Educate Accelerate
Improve Deliver Evaluate Associate
Ignore Decrease Eliminate Avoid

TABLE 4.1 – IDEA

Learning objective: 1. Project feasibility

• Project feasibility

• Find Knowledge gap

• Learn IDEA matrix

• Knowledge canvas

IDEA Matrix:

IDEA matrix is nothing but a matrix representation of characteristic requirement ofthe project.

The IDEA matrix of our project can be thus represented as:

18
 I D E A
Increase efficiency of Drive a search Educate the human to Accelerate speed
Search Engine. Engine which is how to search of Searching
smart enough to be appropriate result result.
search relevant
search.
Improve relevant search Deliver the exact Evaluate technical Associate
result. result of search advancements of Society database with
with help of Smart for its betterment. Inventory system.
crawler.
Ignore irrelevant result. Decrease visiting Eliminate large amount Avoid processing
to unwanted link of processing efforts. in maintaining
of our search daily records of
result. the database.

TABLE 4.2 – IDEA MATRIX

Brief explanation about each characteristic:

Increase: In our project we are thus increase the use and operating efficiency of current search
engine. We are increasing searching capacity of the relevant result. Improve: Improve the traditional
search engine by making it smarter using technologies such as Smart Crawler.

19
 Ignore: We are ignoring the irrelevant result of given searches. Our traditional search engine
gives both results relevant and irrelevant searches among from themwe take relevant search using
smart technologies like smart crawler.
Drive: Hereby we are driving a smart search engine against a traditional search engine which
helps us reducing extra search efforts.
Deliver: We are delivering a quick and easy solution for the maintenance of database that needs
to be updated on regular interval.
Decrease: The extra visit to unwanted result will be decreased by using Smart Crawler and
profession login option also provided on the smart crawler.
Educate: We are trying to make the management authority and efficiency of searchengine
aware of technical advancements around.
Evaluate: By considering the searching on internet reviews and requirements whichneeds to be
satisfied given by the users we are evaluating the technology to be usedalong with algorithms needs to
reduce efforts.
Eliminate: By implementation of smart crawler need for massive number of systemprocessing
is eliminated which leads to efficiency.
Accelerate: Searching is done at much higher speed as there would be we are usingsmart
technologies and algorithms so that it removes unwanted results.
Associate: Here we are associating or linking database with the inventory so that if the sites go
below threshold level inventory must make required arrangements so that the sides should not be
unavailable.
Avoid: If any irrelevant search result in updating database goes may lead to wrong search result
in the system. This needs to be avoided. Hence an updating mechanism is added with help of smart
crawler.

20
KNOWLEDGE CANVAS:

Knowledge canvas is a graphical representation of knowledge gap between any twocomponents of the
project considered.

Knowledge canvas Diagram

Fig 4.1 Knowledge Canvas Diagram

4.2.2. Project problem statement feasibility assessment using NP-Hard, NP-Complete.

P PROBLEM

Polynomial time solving. Problems which can be solved in polynomial time, which take time like O(n),
O(n2), O(n3). E.g.: finding maximum element in an array or to check whether a string is palindrome
or not.
So, there are many problems which can be solved in polynomial time.

NP PROBLEM

Non deterministic Polynomial time solving. An easy example of this is subset sum: given aset of
numbers, does there exist subset whose sum is zero?.

21
But NP problems are checkable in polynomial time means that given a solution of a problem,we can
check that whether the solution is correct or not polynomial.

NP-hard PROBLEM

If a problem is NP-hard, this means I can reduce any problem in NP to that problem. This means if I
can solve that problem, I can easily solve any problem in NP. If we could solve an NP-hard problem
in polynomial time, this would prove P = NP.
NP-complete PROBLEM

A problem is NP-complete if the problem is both NP-hard, and In NP.

Algorithms & Techniques:

Algorithm 1: Exact Pattern Matching AlgorithmO (N + K ).

Complexity: It takes time to fetch URL from web-server, also to extract query entered by user. It

takes data from database as well as from log file.

So, Time Complexity=o(n) OCR-Optical Character Recognition Complexity Analysis

Algorithm 2: OCR-Optical Character RecognitionO (N 2 log (N)).

Overall time required: O(N+K) +O (N 2 log(N))

Space Complexity:

More the storage of data more is the space complexity. Each time we store resultant data in log file

also in database. We store URL (bookmarked) in database. So, moretime complexity.

22
4.3 PROPOSED METHODOLOGY

Admin:

Add Student: The Admin adds the student and the password is generated by thesystem and
sent to the students Mail Id.
Add Course: The Admin is allowed to add the Course and its Subjects semesterwise.
Add Timetable: The Admin is allowed to add the timetable for the course semesterwise in the
form of an .jpg
Add Schedule: The Admin is allowed to add the Schedule for the course semesterwise in the
form of an .jpg
Add Booklet: The Admin adds the booklet limited to a pdf file only.

Add Test Solutions: The Admin adds the test solutions limited to a pdf file only.wise and out
of 25.
Add College related information e.g., Events, workshop doc, photos, branch infowith photos.
Which is useful for represent college.

Student:

Student Login: The Student is allowed to login into the App with password sent tohis/her
email Id and is remembered once logged In.
View Booklet: The Student can see a list of the booklets limited to his/her courseand
semester which are viewed by default by Google docs.
View College related information e.g., Events, workshop doc, photos, branch infowith
photos. Which is useful for represent college.

Parent:

Parent Login: The Parent is allowed to login into the App with password sent tohis/her
email Id and is remembered once logged In

23
4.4 PROJECT TASK SET/PROJECT MANAGEMENT PLAN:

• Task 1-Requirement Gathering, Review of papers

• Task 2-Defining problem statement

• Task 3-Identifying scope and requirements of project

• Task 4-Mathematical analysis

• Task 5-System design analysis

• Task 6-UML diagrams

• Task 7-System Implementation

• Task 8-System Testing

• Task 9-Result Analysis

• Task 10-Documentation

24
 CHAPTER – 5

IMPLEMENTATION DETAILS

5.1 DEVELOPMENT AND DEPLOYMENT SETUP

A college enquiry chatbot can be built using a combination of LSTM (Long Short-Term
Memory) and CNN (Convolutional Neural Network) models to process natural language inputs and
generate appropriate responses.

Here is how it can work:

• Data collection: The first step is to collect a large amount of relevant data, such as frequently
asked questions, course information, admission requirements, campus facilities, etc. This data
will be used to train the chatbotmodel. The relevant data is taken from Concordia university
for the overviewof the project.

• Preprocessing: The first step is to preprocess the text inputs to extract important features and
remove any noise. This can involve steps such as tokenization, stemming, lemmatization, stop
word removal, and spell correction.

Natural Language Processing is a subfield of data science that works with textual data.
When it comes to handling the Human language, textual data is one of the most unstructured
types of data available. NLP is a technique that operates behind the it, allowing for extensive text
preparation prior to any output. Before using the data for analysis in any Machine Learning work, it's
critical to analyse the data. To deal with NLP-based problems, a variety of libraries and algorithms are
employed. For text cleaning, a regular expression(re) is the most often used library. The next libraries
are NLTK (Natural language toolkit) and spacy, which are used to execute natural language
tasks like eliminating stop words.

Pre-processing data is a difficult task. Text pre-processing is done in order to prepare the text
data for model creation. It is the initial stage of any NLP project.

25
5.1.1. REMOVING STOP WORDS

To eliminate noise from data, data cleaning is essential in NLP. Stop words are the most
frequently repeated words in a text that give no useful information. The NLTK library includes a list
of terms that are considered stop words in English. [I, no, nor, me, mine, myself, some, such we, our,
you'd, your, he, ours, ourselves, yours, yourself, yourselves, you, you're, you've, you'll, most, other]
are only a few ofthem.

The NLTK library is a popular library for removing stop words, and it eliminates about 180
stop words. For certain difficulties, we can develop a customized set of stop words.
Using the add technique, we can easily add any new word to a collection of terms. Removing stop
words refers to the process of considered to be common uninformative.

Removing stop words is a common preprocessing step in Natural Language Processing (NLP)
Stop words are words that do not carry much meaning in a sentence or text, such as:

- Articles (the, a, an)

- Conjunctions (and, but, or)
- Prepositions (in, on, at)
- Pronouns (I, you, he, she)
- Auxiliary verbs (is, are, am)

Removing stop words can help improve the performance of NLP models by:

• Improving model efficiency: By removing stop words, the model has to process less data will
can improve its efficiency.

• Enhancing feature extraction: Removing stop words can help the model focus on the more
important words in the text, which can improve feature extraction.

26
5.1.2. LEMMATIZATION

The process of reducing inflected forms of a word while verifying that the reduced form
matches to the language is known as lemmatization. A lemma is a simplified version or base word.
Lemmatization uses a pre-defined dictionary to saves word context and verify the word in the
dictionary as it decreases. Organizes, organized, and organizing, for example, are all forms of organize.
The lemma in this case is organize.The inflection of a word can be used to communicate grammatical
categories such as tense (organized vs organize). Lemmatization is required since it aids in the
reduction of a word's inflected forms into a particular element for analysis. It can also assist in text
normalization and the avoidance of duplicate words with similar meanings.

5.1.3. LOWER CASING

When the text is in the same case, a computer can easily read the words since the machine
treats lower and upper case differently. Words like Cat and cat, for example, are processed differently
by machines. To prevent such issues, we must make the word in the same case, with lower case being
the most preferable instance. In python lower () is a function that is mostly used to handle strings. The
lower () function accepts no parameters. It converts each capital letter to lowercase to produce
lowercased strings from the provided string. If the supplied string has no capital characters, it returns
the exact string.

• Intent Recognition: The next step is to identify the intent behind the user's input. For example,
if the user asks "What are the admission requirements for Computer Science?", the intent can
be recognized as"Admission Requirements".

This can be done using techniques such asrule-based systems, machine learning algorithms
like Naive Bayes, or neural network models like LSTM.

27
• Entity Recognition: Once the intent is recognized, the chatbot needs to extract the relevant
entities from the user's input. In the above example, the entities would be "Computer Science".
This can be done using techniques such as Named Entity Recognition (NER) or Part-of-Speech
(POS) tagging.

• Dialogue Management: The chatbot needs to maintain a conversation flowwith the user and
respond appropriately to their inputs. This can be achieved using techniques such as rule-based
systems, finite-state machines, orreinforcement learning algorithms.

• Response Generation: Finally, the chatbot generates a response to the user's input based on
the intent and entities identified in the previous steps. The response can be a pre-defined
template or a dynamically generated sentence. The response can be generated using techniques
such as rule- based systems, templates, or machine learning algorithms like sequence-to-
sequence models or Generative Pre-trained Transformer (GPT) models.

28
5.2 ALGORITHMS

5.2.1 Long Short-Term Memory (LSTM)

LSTM is a kind of recurrent neural network. In RNN output from the last step is fed as input
in the current step. LSTM was designed by Hochreiter & Schmid Huber. It tackled the problem of
long-term dependencies of RNN in which the RNN cannot predict the word stored in the long-term
memory but can give more accurate predictions from the recent information. As the gap length
increases RNN does not give an efficient performance.

LSTM can by default retain the information for a long period of time. It is used for processing,
predicting, and classifying based on time- series data. Long Short- Term Memory (LSTM) is a type of
Recurrent Neural Network (RNN) that is specifically designed to handle sequential data, such as
timeseries, speech, and text. LSTM networks can learn long-term dependencies in sequential data,
which makes them well suited for tasks such as language translation, speech recognition, and time
series forecasting. A traditional RNN has a single hidden state that is passed through time, which can
make it difficult for the network to learn long-term dependencies. LSTMs address this problem by
introducing a memory cell, which is a container that can hold information for an extended period.

The memory cell is controlled by three gates: the input gate, the forget gate, and the output
gate. These gates decide what information to add to, remove from, and output from the memory cell.
The input gate controls what information is added to the memory cell.

The forget gate controls what information is removed from the memory cell. And the output
gate controls what information is output from the memory cell. This allows LSTM networks to
selectively retain or discard information as it flows through the network, which allows them to learn
long-term dependencies.

29
 Fig 5.1 Structure of LSTM

LSTMs can be stacked to create deep LSTM networks, which can learn even more complex
patterns in sequential data. LSTMs can also be used in combination with other neural network
architectures, such as Convolutional Neural Networks (CNNs) for image and video analysis.LSTM
has a chain structure that contains four neural networks and different memory blocks called cells.
Information is retained by the cells and the memory manipulations are done by the gates.

30
There are three gates –

1. Forget Gate: The information that is no longer useful in the cell state is
removed with the forget gate. Two inputs x_t (input at the particular time) and h_t-1 (previous cell
output) are fed to the gate and multiplied with weight matrices followed by the addition of bias. The
resultant is passed through an activation function which gives a binary output. If for a particular cell
state the output is 0, the piece of information is forgotten and for output 1, the information is retained
for future use.

Fig 5.2 Forget Gate

2. Input gate: The addition of useful information to the cell state is done by the
inputgate. First

Fig 5.3 Input Gate

31
The information is regulated using the sigmoid function and filter the values to be remembered like
the forget gate using inputs h_t-1 and x_t.

Then, a vector is created using tanh function that gives an output from -1 to +1, which contains all the
possible values from h_t-1 and x_t. At last, the values of the vector and the regulated values are
multiplied to obtain the useful information

3. Output gate: The task of extracting useful information from the current cell state to be
presented as output is done by the output gate. First, a vector is generated by applying tanh
function on the cell. Then, the information is regulated using the sigmoid function and filter
by the values to be remembered using inputs h_t-1 and x_t. At last, the values of the vector
and the regulated values are multiplied to be sent as an output and input to the next cell

Fig 5.4 Output Gate

.

32
5.2.2 CONVOLUTIONAL NEURAL NETWORK (CNN)

A Convolutional Neural Network (CNN) is a type of deep learning algorithm commonly used
in image recognition and computer vision applications. This stands for Convolution Neural Network
where Image data is mapped to a target variable. They have proven to be successful in that they are
now the techniques of choice forany form of prediction issue utilizing data as an input to the model.
CNN is a multi- layered feed-forward neural network that is built by layering several hidden layers on
top of one another in a certain sequence. These layers are frequently outlawed by several layers in
CNN, while activation layers are usually enhanced by layers in the convolutional network. In the
context of a college enquiry chatbot system, CNN can be useful in several ways:

Image Recognition: CNN can help the chatbot to identify images related to collegeenquiries.
For example, if a user sends an image of a college campus, the chatbot can use a pre-trained CNN
model to recognize the image and extract relevant information such as the name of the college, its
location, and other details that canassist the user in their enquiry.

Data Analysis: CNN can be used to analyze textual data related to college enquiries. For
example, if a user asks a question about admission requirements for a particular program, the chatbot
can use a CNN model to extract the most importantkeywords and concepts from the text and provide
a relevant response based on thatinformation.

Improved Accuracy: Using a CNN model can improve the accuracy of the chatbot'sresponses,
as it can quickly and accurately analyze large amounts of data related to college enquiries and provide
the most relevant responses to users.

Chatbot training: CNNs can be used as a part of the training process for chatbots. For
example, CNNs can be used to analyze large datasets of user queries and responses to identify patterns
and improve the chatbot's ability to understand and respond to user queries.

33
 Text classification: CNNs can be used to classify user input into different categories or intents.
This is useful in chatbots as it allows the chatbot to understand the user's query and respond
appropriately. CNNs can learn to identify patterns in text data and can be trained on large datasets to
improve their accuracy.

Entity extraction: CNNs can be used to extract relevant information from unstructured text
data, such as course descriptions or faculty biographies. This canbe useful in chatbots for providing
detailed information to users.

Contextual understanding: CNNs can be used to improve the chatbot's contextual

understanding of user input.

Overall, CNN can be a valuable tool in a college enquiry chatbot system, as it can help to
enhance the accuracy and effectiveness of the chatbot in responding to user enquiries, especially when
it comes to analyzing visual and textual information. Contextual understanding refers to the ability to
coin the context in which it is presented.

34
5.3 MODULE IMPLEMENTATION
5.3.1. RDFLIB

RDFLib is a pure Python package for working with RDF. RDFLib contains mostthings you
need to work with RDF, including:
• Parsers and serializers for RDF/XML, N3, NTriples, N-Quads, Turtle, TriX,Trig and JSON-
LD
• A Graph interface which can be backed by any one of a number of Storeimplementations
• Store implementations for in-memory, persistent on disk (Berkeley DB) andremote SPARQL
endpoints
• A SPARQL 1.1 implementation - supporting SPARQL 1.1 Queries andUpdate statements
• SPARQL function extension mechanisms

5.3.2. RE

A RegEx, also known as a Regular Expression, is a string of characters that definesa search
patterns. This module's functions allow to see if a given string matches a given regular expression.

5.3.3. RANDOM

The Python Random module is a built-in module for generating random integers in Python.
These are sort of fake random numbers which do not possess true randomness. We can therefore use
this module to generate random numbers, display a random item for a list or string, and so on.

35
5.3.4. CSV
The CSV module implements classes to read and write tabular data in CSV format. It allows
programmers to say, “write this data in the format preferred by Excel,” or “read data from this file
which was generated by Excel,” without knowing the precisedetails of the CSV format used by Excel.
Programmers can also describe the CSV formats understood by other applications or define their own
special- purpose CSV formats.

5.3.5. SPOTLIGHT
Data validation for Python, inspired by the Laravel framework. The main focus of the
Spotlight library is on deep learning techniques such as matrix factorization, neural networks, and
sequence modeling. It provides a flexible framework for building different types of recommendation
systems, including collaborative filtering, content-based filtering, and hybrid models.

36
5.4 DATA FLOW DIAGRAMS

Fig 5.5.1 Level 0 Data Flow Diagram

37
 Fig 5.5.2 Level 1 Data Flow Diagram

A data flow diagram (DFD) is a graphical representation of the flow of data in a system. In the
context of the chatbot system for college enquiry using a knowledgeable database, a DFD can be used
to illustrate the flow of data between the various components of the system. The DFD can help in
understanding the datainputs, processing, and outputs of the system. The DFD for the chatbot system
can be divided into four main components: the user interface, the natural language processing engine,
the knowledgeable database, and the response generation component. The user interface component
receives the input queries from the user in natural language. The input query is then passed on to the
natural language processing engine, which processes the query and extracts relevant information such
as intent, entities, and sentiment.

38
5.5 USE CASE DIAGRAM

Fig 5.6 Use Case Diagram

A use case diagram is a graphical representation of the interactions between actors(users) and
the system. In the context of the chatbot system for college enquiry using a knowledgeable database,
a use case diagram can be used to identify the various use cases or scenarios in which the system is
used.The use case diagram for the chatbot system can include the actors (users) such as prospective
students, parents, and other stakeholders who are interested in obtaining information about the college.
The various use cases can include querying information about courses,admission requirements, campus
facilities, and other related information.

39
5.6 CLASS DIAGRAM

Fig 5.7 Class Diagram

A class diagram is a type of UML (Unified Modeling Language) diagram that represents the
classes and their relationships in a system. In the context of the chatbot system for college enquiry
using a knowledgeable database, a class diagram can be used to represent the various classes in the
system and their relationships.The class diagram for the chatbot system can include classes such as
User, Query, Response, Natural Language Processing Engine, Knowledgeable Database, Retrieval-
based Algorithm, Rule-based Algorithm, Machine Learning Algorithm, Hybrid Approaches, and
Feedback Mechanism. Each class can have attributes and methods that define its behavior and
properties.

40
5.7 SEQUENCE DIAGRAM

Fig 5.8 Sequence Diagram

A sequence diagram is a type of UML (Unified Modeling Language) diagram that represents
the interactions between objects in a system over time. In the context of the chatbot system for college
enquiry using a knowledgeable database, a sequence diagram can be used to represent the sequence of
interactions between the user and the system when making a query. The sequence diagram for the
chatbot system can include the user object, query object, response object, and the various algorithms
and components of the system such as the Natural Language Processing Engine, Knowledgeable
Database, Retrieval-based Algorithm, Rule- based Algorithm, Machine Learning Algorithm, Hybrid
Approaches, and Feedback Mechanism.

41
5.8 COMPONENT DIAGRAM

Fig 5.9 Component Diagram

A component diagram is a type of UML (Unified Modeling Language) diagram that represents
the physical and logical components of a system and their relationships. In the context of the chatbot
system for college enquiry using a knowledgeable database, a component diagram can be used to
represent the various components of the system and their relationships. The component diagram for
the chatbot system can include components such as User Interface, Natural Language Processing
Engine, Knowledgeable Database, Retrieval-based Algorithm, Rule- based Algorithm, Machine
Learning Algorithm, Hybrid Approaches, and Feedback Mechanism. Each component can have its
own set of interfaces, ports, and dependencies that define its behavior and interactions with other
components.

42
5.9 DEPLOYMENT DIAGRAM

Fig 5.10 Deployment Diagram

A deployment diagram is a type of UML (Unified Modeling Language) diagram that represents
the physical deployment of components and their relationships in a system. In the context of the chatbot
system for college enquiry using a knowledgeable database, a deployment diagram can be used to
represent the physical deployment of the various components of the system. The deployment diagram
for the chatbot system can include nodes such as User Interface Node, Natural Language Processing
Engine Node, Knowledgeable Database Node, and Feedback Mechanism Node. Each node can
represent a physical machine or a logical grouping of machines that host the corresponding
components.

43
5.10 COLLABORATION DIAGRAM

Fig 5.11 Collaboration Diagram

44
5.11 STATE CHART DIAGRAM

Fig 5.12 State Chart Diagram

45
 CHAPTER-6

RESULTS AND DISCUSSION

Fig.6.1. Execution ( Output )

46
 Fig.6.2. Execution ( Output )

Our Chatbot provides information regarding to the college. It is the website. It is communicate
to the client like guardians, understudy. By utilizing NLP human language changed into an information
language. By utilizing AI to client give collegedata. This could be type-based (composed) discussion,
even a non-verbal discussion. At the point when ChatBot innovation is incorporated with well known
web administrations it very well may be used safely by a significantly bigger crowd. Chabot framework
is carried out to meet scholarly necessities of the clients. Generating reaction from a Chabot is
information Word Net is answerable for recovering the reactions and for this situation; it contains all
rationales that is set off at whatever point the client setting is coordinated. At the point when a client
starts asking questions in the Chabot Graphical User Interface (GUI). The question is looked in the
information base..On the off chance that the reaction is found in the information base it is shown to
the client else the framework tells the administrator about the missing reaction in the data set and gives
a predefined reaction to the client.

47
 Several studies have been conducted on college enquiry chatbots, and the results suggest that
chatbots can significantly improve the efficiency and effectiveness of college enquiries.
Here are some brief results and discussions from these studies:

Improved efficiency: Chatbots can significantly reduce the workload on college

administrators and provide faster, more efficient, and personalized services to students. For example,
a study by Turel et al. (2021) found that a chatbot developed for student admissions reduced the average
response time from 3 days to less than1 minute.

Higher user satisfaction: Several studies have found that students are generally satisfied with
the performance of college enquiry chatbots. For example, a study byStieger et al. (2020) found that
students rated the chatbot developed for their university highly on ease of use, usefulness, and overall
satisfaction.

Accuracy and effectiveness: Chatbots have been shown to be effective in handling a wide
range of college enquiries, including admission inquiries, course registration, financial aid, campus
facilities, and career services. However, accuracyand effectiveness can vary depending on the quality
of the chatbot's NLP and ML algorithms.

Challenges in chatbot development: Developing an effective college enquiry chatbot

is not without challenges. Challenges include accurately identifying user intent, managing a large
knowledge base, providing multilingual support, maintaining context across conversations, and
ensuring a positive user experience.

Future research directions: Several research directions have been proposed for college
enquiry chatbots, including improving NLP and ML algorithms, designing chatbots that can handle
complex and multi-turn conversations, providing personalized recommendations and support, and
developing chatbots that can handle emotional and mental health inquiries.
Overall, the results and discussions from the literature suggest that college enquiry chatbots ,
providing faster and personalized services to students. However, there is still much work to be done to
improve the accuracy and effectiveness of chatbots and to address the challenges in chatbot
development. Chatbots can gather data onuser queries, preferences, and behavior, which can be used
to improve the chatbot's performance and inform college decision- making.

48
 Chatbots can provide a more conversational and This can lead to increased user engagement
and satisfaction. Chatbots can handle routine and repetitive enquiries, freeing up staff time to focus on
more complex queries and tasks.Chatbots can be accessed anytime and anywhere through a range of
devices, making it easier for students to get the information they need. Chatbots can be designed to
provide personalized responses based on the user's profile, interests, and previous interactions with the
chatbot.

Chatbots can handle multiple enquiries simultaneously, providing quick and efficient responses
to users. However, there are also some challenges and limitations to the implementation of college
enquiry chatbot.

49
 CONCLUSION

7.1 CONCLUSION

Fastest-growing technology in history is artificial intelligence. utilizing a database that is both

artificially intelligent and knowledgeable. We are able to transform virtual aid and pattern matching.
This method is creating a chatbot based on the Android operating system with the help of a virtual
assistant and an artificially intelligent database. A chatbot that can distinguish between human and
machine speech andanswers to user enquiries is something we can make. Researchers must
cooperateand decide on a common strategy in order to build a chatbot. In this study, we investigated
the development of chatbots and their applications across several industries. Also, there are parallels
with other chatbots. The knowledge base of the chatbot should generally be brief, approachable, and
simple to understand. Even if some of the commercial solutions have just become accessible, there is
still work tobe done in order to discover a standard method for building chatbots. In conclusion, a
chatbot system for college enquiry using a knowledgeable database can provide a convenient and
efficient way for students, faculty, and other stakeholders to access information about college
programs, courses, and admission requirements. By utilizing natural language processing algorithms,
a knowledgeable database, and a recommendation engine, the chatbot system can generate accurate
and relevant responses to user queries in a timely manner. Save timing of students and teachers and
also save extra manpower. Student can see all document related college like, notice, study material,
question papers etc. on time to time and from any place. It is proper communication in between staff
and students.

50
7.2 FUTURE WORK

As stated in the paper, the project has a broad reach in the current context. The proposal's
majority of proposed features have been implemented. So, if I continue working on this project, I
intend to create a database for the system where the admi may keep the extracted data. Further, future
study will include a more in- depth examination of certain techniques, further research on other
libraries, and new approaches to explore different methods.

Integration with other systems: Integrate with student information systems, CRM, or ERP
systems for seamless data exchange.

Personalization: Use machine learning to offer personalized college recommendations based

on users' interests, academic background, and career goals.

Multilingual support: Add support for multiple languages to cater to a broader audience.

Voice-based interface: Develop a voice-based interface to interact with the chatbot using voice
commands.

51
 APPENDIX

A. SOURCECODE

from flask import Flask, render_template, request, jsonifyimport random

import json import pickle

import numpy as npimport nltk

from nltk.stem import WordNetLemmatizer

from tensorflow.keras.models import load_model

app = Flask( name )

# Initialize lemmatizer

lemmatizer = WordNetLemmatizer()

# Load intents and model

intents=json.loads(open('intents.json').read())

word = pickle.load(open('words.pkl', 'rb'))

classes= pickle.load(open('classes.pkl', 'rb'))

model= load_model('chatbotmodel.h5')

# Function to clean up and tokenize sentence def clean_up_sentence(sentence):

sentence_words = nltk.word_tokenize(sentence)

sentence_words = [lemmatizer.lemmatize(word.lower()) for word insentence_words]

return sentence_words

# Function to create a bag of wordsdef bag_of_words(sentence):

sentence_words = clean_up_sentence(sentence)bag = [0] * len(words)

for w in sentence_words:

for i, word in enumerate(words):

if word == w:

52
bag[i] = 1

return np.array(bag)

# Function to predict class def predict_class(sentence):

bow = bag_of_words(sentence)

res = model.predict(np.array([bow]))[0]

ERROR_THRESHOLD = 0.25

results = [[i, r] for i, r in enumerate(res) if r > ERROR_THRESHOLD] results.sort(key=lambda x:

x[1], reverse=True)

return_list = [{'intent': classes[r[0]], 'probability': str(r[1])} for r in results]return return_list

# Function to get a response def get_response(intents_list, intents_json):if intents_list:

tag = intents_list[0]['intent'] for

53
i in intents_json['intents']: if i['tag'] == tag:
return random.choice(i['responses'])return "Sorry, I didn't understand that."

@app.route('/'

)def index():

return render_template('index.html')

@app.route('/get-response', methods=['POST'])def get_bot_response():

user_message = request.json.get('message')ints = predict_class(user_message) bot_response
= get_response(ints, intents) return jsonify({'response': bot_response})

if name == ' main

':

app.run(debug=True) import json

import pickle

import numpy as npimport nltk nltk.download('punkt')

from nltk.stem import WordNetLemmatizer from

tensorflow.keras.models import Sequential from

tensorflow.keras.layers import Dense

Dropoutfrom tensorflow.keras.optimizers import SGD

# Initialize lemmatizer

lemmatizer = WordNetLemmatizer()

54
# Load data

with open('intents.json') as file:intents = json.load(file)

# Load pickle files

with open('words.pkl', 'rb') as file:words = pickle.load(file)

with open('classes.pkl', 'rb') as file:classes = pickle.load(file)

# Create training data def tokenize(sentence):

return [lemmatizer.lemmatize(word.lower()) for word innltk.word_tokenize(sentence)]

def bag_of_words(sentence): sentence_words = tokenize(sentence)bag = [0] * len(words)

for word in sentence_words:if word in words:
bag[words.index(word)]

= 1return np.array(bag)
X_train = []y_train
= []

for intent in intents['intents']: for pattern in

intent['patterns']: bag = bag_of_words(pattern) X_train.append(bag)
y_train.append(classes.index(intent['tag']))

X_train = np.array(X_train)y_
train
= np.array(y_train)

# Build and train the modelmodel = Sequential()

model.add(Dense(128, input_shape=(len(X_train[0]),), activation='relu'))model.add(Dropout(0.5))

model.add(Dense(len(classes), activation='softmax'))

55
model.compile(loss='sparse_categorical_crossentropy', optimizer=SGD(learning_rate=0.01),
metrics=['accuracy'])

model.fit(X_train, y_train, epochs=200, batch_size=5, verbose=1)

# Save the model model.save('chatbotmodel.h5')

print("Model trained and saved as 'chatbotmodel.h5'.")

56
{"intents": [
{"tag": "greetings",
"patterns": ["hello","hey","hi","good day","Greetings","what's up?","how is it going?"],
"responses": ["Hello!","Hey!","What can I do for you?"]
},

{"tag": "name",
"patterns": ["what is your name","name","what's your name","who are you","what should I call
you"],
"responses": ["You can call me hitsbot","I'm hitsbot","I'm hitsbot your virtual assistant"]
},

{"tag": "courses",
"patterns": ["what courses are available", "how many courses are there in this college"],
"responses": ["For Bachelors in Business Administration we have been offering the
followings:\n\n1.Bachelors in Technology we have been offering the
followings:\n\n1.CSE\n\n2.MECH\n\n3.ECE\n\n4.EEE\n\n6.ARCH\n\n7.CIVIL"]
},

{"tag": "courseDuration",
"patterns": ["how long will be B.TECH course", "how long will it take to complete BBA or B.TECH
course"],
"responses": ["Our college offers 4 year long B.TECH course."]
},

{"tag": "Location",
"patterns": ["location","where is it located","what is the location of the college"],
"responses": ["K.Ramakrishnan college of technology ,Samapuram ,trichy."]
,

57
{"tag": "studentRequirements",
"patterns": ["what are the student requirements for admission","entry requirements","admission
requirements"],
"responses": ["Academic Level\nHSC +2 overall 55% ."]
},

{"tag": "classes",
"patterns": ["how many classes will be there in a day","how long are the classes?"],
"responses": ["There may be seven classes per day. Each class will be of 50 minutes."]
},

{"tag": "teachingStyle",
"patterns": ["what is the teaching style of this college?","Is the teaching pattern different from other
college?","what is the teaching format?"],
"responses": ["Our college has different teaching patterns than other colleges of tamilnadu.\nYou can
provide us with your contact details and our counselors shall reach out to you and provide you with
further details."]
},

{"tag": "exams",
"patterns": ["what are the exams like?","What is the exam pattern"], "responses": ["There are
assignments which carry more weight than your written
exams. The assignments have deadlines which you should not exceed if you want toget better
marks."]
},

{"tag": "hours",
"patterns": ["what are your hours","when are you guys open","what your hours of operation"],
"responses": ["You can message us here at any hours. But our college premises will be open from
8:45 am to 4:30 pm only."]
},

58
A. SCREENSHOTS

Fig.B.1. Execution ( Output )

Fig.B.2. Execution ( Output )

59
 REFERENCES

[1] CHATBOT BASED ON AI Information technology professors Nikita Hatwarl, Ashwini

Patil 2, and Diksha Gondane 3123 are from Nagpur/RTMNU in India. Volume 3, Issue 2 (March-April
2016), International Journal of Emerging Trends in Engineering and Basic Sciences (UEEBS),
ISSN(Online)2349-6967)

[2] Chatbot Utilizing a Knowledge in Database at the 2016 7th InternationalConference on

Intelligent Systems, Modelling and Simulation by Bayu Setiaji and Ferry Wahyu Wibowo.

[3] Chatbot, https://en.wikipedia.org/wiki/Chatbot.Schantz, Herbert F., The History of

OCR, Recognition Technologies Users Association, ManchesterCenter, Vermont, 1982.

[4] P. Chenna, T. Rao, et al (2018). HFSS software is used to design a twin inverted L
microstrip antenna for sustainable systems. International Journal of Computer-Aided Engineering
Technology (IJCAET).

[5] J. Quintero, an IEEE student member, and R. Asprilla, an IEEE memberProceedings of

the 2020 IEEE Thirty-Fifth Central American and Panama Convention: Towards an Efficient Voice-
Based Chatbot (concapan xxxv).

[6] Rao, P.T., and Kumar, P.P. (2015,January). Twin microstrip patch antenna in the form
of a staircase. 2015 saw the ICPC (International Conference on Pervasive Computing) (pp. 1-5). The
Technical Writer's Handbook by IEEE
M. Young. University Science, Mill Valley, California, 1989.

[7] Rao, P. T., and Kumar, P. P. (2021). dual dual folded inverted-L antennawith a
frequency-tunable circular polarisation and varactor-loaded split-ring resonator constructions.
Global Journal of Communication Systems, e4715.

[8] Ly Pichponreay and Chi-Hwan Choi created the Chungbuk National University
Smart Responding Chatbot based on OCR and Overgenerating Transformations and Rating.

[9] Trinatha Rao, P., and Praveen Kumar, P. C. (2020). Using HFSS, a double folded
inverted-L antenna with a rectangular ground plane is designed to be reconfigurable.
IETNetworks,9(5),229-234.

[10] Automated Question Answering System Utilizing Ontology and Semantic Role
International Conference on Innovative Mechanisms for Industrial Applications (ICIMIA2017) by S.
Jayalakshmi and Dr. Ananthi Sheshasaayee.

60
[11] Yuriy Dyachenko and Nayden Nenkov Articial Intelligence Technologies For Personnel
Learning Management Systems 2019 IEEE 8th Internatignal Conference on Intelligent Systems
Volodymyr DahlEast Ukrainian National University Severodonetsk, Ukrainev.

[12] "Development of Chatbot System for College Enquiry" by J. Rajitha and

S. Sathish Kumar. International Journal of Engineering Research & Technology,vol. 6, no. 3,

March 2017.

[13] "Chatbot Based Enquiry System for College" by S. Sowmiya and S. Sathish Kumar.
International Journal of Innovative Research in Science, Engineering and Technology, vol. 6, no. 8,
August 2017.

[14] "Development of Chatbot for College Enquiry System using AIML" by R. Arun and A.
Shanthi. International Journal of Computer Science and Mobile Computing, vol. 6, no. 9, September
2017.

[15] "Chatbot Based Enquiry System for College Admissions" by S. S. Gupta and R. K. Singh.
International Journal of Computer Sciences and Engineering, vol. 6, no. 2, February 2018.

[16] "An AI-Based Chatbot System for College Enquiry" by N. Nidheesh et al.(2020). This
paper describes the development of an AI-based chatbot system for college enquiry that utilizes a
knowledgeable database and machine learningtechniques to provide accurate responses to user queries.

61