Download as pdf or txt
Download as pdf or txt
You are on page 1of 8

VISVESVARAYA TECHNOLOGICAL UNIVERSITY

“Jnana Sangama”, Belagavi-590018

Major Project: Phase-II Report On

“Cyberbully Detection in Twitter”

18CSP83
Submitted By
Divyashree M L (1RR20IS010)
Gouthami V N (1RR20IS011)
Nanditha R (1RR20IS026)
Under the Guidance of
Dr. J Amutharaj
Professor and HOD
Dept. of ISE
Rajarajeswari College of Engineering

Submitted in partial fulfillment for the award of degree of


BACHELOR OF ENGINEERING
in
INFORMATION SCIENCE AND ENGINEERING

DEPARTMENT OF INFORMATION SCIENCE & ENGINEERING


RAJARAJESWARI COLLEGE OF ENGINEERING
MYSORE ROAD, BANGALORE – 560074
(An ISO 9001-2008 Certified Institute)
2023-2024
RAJARAJESWARI COLLEGE OF ENGINEERING
Kumbalgodu, Mysore Road, Bengaluru-560074
(An ISO 9001:2008 Certified Institute)
DEPARTMENT OF INFORMATION SCIENCE & ENGINEERING

CERTIFICATE

Certified that the project work titled “Cyberbully Detection in Twitter” carried out by
Divyashree M L (1RR20IS010), Gouthami V N (1RR20IS011), and Nanditha R (1RR20IS026) are
bonafide students submitted in partial fulfillment for the award of degree of Bachelor of Engineering
in Information Science and Engineering of “Rajarajeswari College of Engineering”, Bengaluru
affiliated to Visvesvaraya Technological University, Belagavi during the year 2023-24. It is certified
that all corrections indicated for internal assessment have been incorporated in the report. The above
project report has been approved as it satisfies the academic requirement in respect with project work
prescribed for the said degree.

………………………… ………………………… …………………………


Signature of Guide Signature of HOD Signature of Principal
Dr. J. Amutharaj Dr. J. Amutharaj Dr. R. Balakrishna
Professor & HOD Professor & HOD Professor & Principal
Dept. of ISE Dept. of ISE RRCE, Bengaluru
RRCE, Bengaluru RRCE, Bengaluru

Name of the Examiners Signature with Date

1)…………………………… ……………………....

2)............................................ ...…………………….
ACKNOWLEDGEMENT

We take this opportunity to thank a lot of eminent personalities without whose constant encouragement
this endeavour of ours would not have become a reality. We consider ourselves privileged to express
gratitude and respect towards all those who guided us throughout our work.

We would like to thank Visvesvaraya Technological University, Belagavi for having this project in
curriculum, which gave us the wonderful opportunity to work on our research and presentation
abilities.

We express our sincere gratitude to Dr.R Balakrishna, Principal, Rajarajeswari College of


Engineering for providing us congenial environment and surroundings to work.

We express our sincere thanks to our Internal Guide Dr. J Amutharaj, Vice Principal, Professor and
HOD of Information Science and Engineering, Rajarajeswari College of Engineering, for his able
guidance, valuable advice and support during this project work.

We extend our sincere thanks to the teaching and non-teaching staff members of our ISE department
for their kindness forward to us in completing the project work.

DIVYASHREE M L (1RR20IS010)

GOUTHAMI V N (1RR20IS011)

NANDITHA R (1RR20IS026)

i
ABSTRACT

The pervasive nature of hate speech and cyberbullying on social media platforms especially in twitter
underscores the critical need for automated detection mechanisms to mitigate their harmful effects on
society, particularly on marginalized groups. This project addresses this demanding issue by proposing
an automated system capable of identifying social media bullying through a comprehensive analysis of
textual and visual content, particularly focusing on tweets. Leveraging advanced techniques such as
Natural Language Processing (NLP) and Convolutional Neural Networks (CNN), the model aims to
accurately detect instances of bullying behavior.

Through the utilization of NLP, the model effectively conducts word similarity analysis in tweets,
enabling the identification of patterns indicative of bullying behavior. Additionally, CNNs are employed
for image recognition, further enhancing the model's capability to detect bullying instances embedded
within multimedia content. This dual approach ensures a robust and comprehensive detection
mechanism, capable of handling diverse forms of cyberbullying prevalent on social media platforms like
Twitter.

Furthermore, the project integrates the use of machine learning algorithms, including XGBoost, Random
Forest, and Support Vector Classification (SVC), to develop a sophisticated cyberbullying detection
model. By preprocessing textual data and extracting features such as sentiment and linguistic patterns,
the model gains insights into the nuanced nature of cyberbullying behavior. Implemented in Python,
these models undergo rigorous training to learn complex relationships between features and
cyberbullying instances, thereby enhancing their effectiveness in identifying and addressing
cyberbullying.

Evaluation of the proposed model is conducted using robust metrics such as accuracy and cross-
validation to assess its performance and generalizability. The results demonstrate the model's efficacy
in detecting cyberbullying instances across various social media platforms, providing a valuable
framework for combating cyberbullying in online environments. Ultimately, this project contributes to
the development of scalable and effective solutions for mitigating the harmful impacts of cyberbullying,
thereby fostering a safer and more inclusive online community.

ii
TABLE OF CONTENTS

Acknowledgement i
Abstract ii
Table of Contents iii,iv,v
List of Figures v
List of Tables vi
Chapter 1 1-9
Introduction
1.1 Overview of the project 1-2
1.2 Details of the project 2-3
1.2.1 Literature Review 2-5
1.2.2 Problem Statement 5
1.2.3 Objective 5-6
1.2.4 Scope 6
1.2.5 Methodology 7-8
1.2.6 Organization of Report 8-9
Chapter 2 10-13
Theory and concept
2.1 Concepts related to the domain of work 10-11
2.2 System Models 11-12
2.3 Application 13
2.4 Summary 13
Chapter 3 14-18

Software Requirement Specification

3.1 Overall Description


3.1.1 Product Perspective 14
3.1.2 Product Functions 14
3.2 Specific Requirements 14-15
3.2.1 Functional Requirements 15
3.2.2 Performance Requirements 15
3.2.3 Hardware Requirements 15
3.2.4 Software Requirements 16
iii
3.2.5 Non-Functional Requirements 16-17
3.2.6 Design Constraints 17-18
3.3 Summary
Chapter 4 19-28
High Level Design
4.1 Design considerations 19
4.1.1 General Design Considerations 19
4.1.2 Development Method 19-21
4.2 Architecture Strategies 21-23
4.2.1 Programming language 21
4.2.2 System Architecture 21-23
4.3 Dataflow Diagrams 23-28
4.3.1 Dataflow Level 0 23-24
4.3.2 Dataflow Level 1 24-25
4.3.3 Dataflow Level 2 26-27
4.4 Summary 28
Chapter 5 29-30
Detailed Design
5.1 Structure Chart 29-30
5.2 Summary 30
Chapter 6 31-33
Implementation
6.1 Programming Language selection 31-32
6.2 Code convention 32-33
6.3 Summary 33
Chapter 7 34-39
Software testing
7.1 Test Environment 34-35
7.1.1 Testing Methods 35
7.1.2 Levels of Testing 35-39
7.2 Unit Testing 36-37
7.3 Integration Testing 37-38
7.4 System testing 38-39
7.5 Summary 39

iv
Chapter 8 40-44
Experimental Analysis and Results
8.1 Performance Analysis
8.2 Result
Chapter 9 45-46
Conclusion 45
9.1 Limitations 45-46
9.2 Future Enhancement 46
References 47-48
Appendices 49

List of Figures
Figure Name Page No.

Fig 1.1: Methodology for Cyberbullying Detection 7


Fig 4.2 System Architecture 21
Fig 4.2 Dataflow Diagram Level 0 24
Fig 4.3 Dataflow Diagram Level 1 25
Fig 4.4 Dataflow Diagram Level 2 27

Fig 5.1 Structure Chart 29


Fig 8.1 User Register Page (Sign up page) 41

Fig 8.2 User Login Page (Sign in page) 41

Fig 8.3 Front Page of UI 42

Fig 8.4: User blocked from login into his account as the reputation score became < 5 42

Fig 8.5 Profile page of Blocked user 43

Fig 8.6 Search Page 43

Fig 8.7 Searched user page 44

Fig 8.8 Profile with reputation score above 9 and prototype of follow 44

Fig 9.2 Accuracy of ML models 49

Fig 9.3 Accuracy of LSTM 49

v
List of Tables

Tables Page No.


Table:7.2 Unit Testing 37

Table:7.3 Integrating Testing 38

Table:7.4 System Testing 39

Table:8.1 Model and Accuracy 40

vi

You might also like