A

PROJECT REPORT
ON

“A HYBRID DEEP LEARNING APPROACH FOR

CYBERBULLYING DETECTION IN TWITTER SOCIAL MEDIA
PLATFORM”
SUBMITTED IN PARTIAL FULLFILLMENT OF THE REQUIREMENT FOR THE AWARD OF THE
DEGREE OF

BACHELOR OF TECHNOLOGY
IN
COMPUTER SCIENCE AND ENGINEERING
SUBMITTED BY

B.Uma devi - 196M1A0572

V.Sruthi - 196M1A0593
N.Alekya - 196M1A05B5
A. Rupa bindu -196M1A05B6
B. Sravani -206M5AO502
Under the guidance of
. Ms. Sanku Vasavi MTech

COMPUTER SCIENCE AND ENGINEERING

PRINCETON INSTITUTE OF ENGINEERING & TECHNOLOGY
FOR WOMEN
(Affiliated to JNTUH & Approved by AICTE)
Chowdaryguda(V), Ghatkesar(M), Medchal, Malkajgiri Dist TS-
500088
A
PROJECT REPORT
ON

“A HYBRID DEEP LEARNING APPROACH FOR

CYBERBULLYING DETECTION IN TWITTER SOCIAL MEDIA
PLATFORM”

SUBMITTED IN PARTIAL FULLFILLMENT OF THE REQUIREMENT FOR THE AWARD OF THE

DEGREE OF

BACHELOR OF TECHNOLOGY
IN
COMPUTER SCIENCE AND ENGINEERING
SUBMITTED BY

B .Uma devi - 196M1A0572

V.Sruthi - 196M1A0593
N.Alekya - 196M1A05B5
A. Rupa bindu -196M1A05B6
B. Sravani -206M5AO502
Under the guidance of
Ms.Sanku Vasavi, MTech

COMPUTER SCIENCE AND ENGINEERING

PRINCETON INSTITUTE OF ENGINEERING & TECHNOLOGY
FOR WOMEN
(Affiliated to JNTUH & Approved by AICTE)
Chowdaryguda(V), Ghatkesar(M), Medchal, Malkajgiri Dist TS-
500088
PRINCETON INSTITUTE OF ENGINEERING & TECHNOLOGY
FOR WOMEN
(Affiliated to JNTUH & Approved by AICTE)

(Chowdaryguda(V), Ghatkesar(M), Medchal, Malkajgiri Dist TS-

500088)

DEPARTMENT OF COMPUTER SCIENCE AND ENGINEERING

CERTIFICATE

This is to certify that the project report titled “A HYBRID DEEP LEARNING
APPROACH FOR CYBERBULLYING DETECTION IN TWITTER SOCIAL
MEDIA PLATFORM” is submitted by B.UMA DEVI (196M1A0572),V.SRUTHI
(196M1A0593),N.ALEKY( 196M1A05B5) ,A .RUPA BINDU
(196M1A05B6),B.SRAVANI (206M5A0502), in B.Tech 4 2 semester COMPUTER
SCIENCE AND ENGINEERING is a record bonofide work carried out by them.
The results embodied in this report have not been submitted to any other university
for the award of any degree.or prize.

INTERNAL GUIDE HEAD OF THE DEPARTMENT

Ms.SANKU VASAVI MTech Mrs. SWAPNA INDRAPALLA MTech

Associate professor

Viva voce held on :

PRINCIPAL
EXTERNAL EXAMINER Dr.RAJEER SHRIVASTAVA
MTech,Ph.D
DECLARATION

We , the students of PRINCETON INSTITUTE OF ENGINEERING AND

TECHNOLOGY FOR WOMEN hereby declare that this is work entitled “A
HYBRID DEEP LEARNING APPROACH FOR
CYBERBULLYING DETECTION IN TWITTER SOCIAL MEDIA
PLATFORM ” , being submitted to the Department of CSE, PRINCETON
COLLEGE affiliated to JNTUH, Hyderabad for the award of BACHELOR OF
TECHNOLOGY in Computer Science and Engineering is a record of University
for the award of any other degree or prize.

B.Uma devi - 196M1A0572

V.Sruthi - 196M1A0593
N.Alekya - 196M1A05B5
A. Rupa bindu -196M1A05B6
B. Sravani -206M5AO502
ACKNOWLEGEMENT

The present project work in the several days study of the various aspects of the
project development. During this the effort in the present study. We have received a
great amount of help from out Chairman Dr. S. PRABAKAR RAO which we wish
to knowledge and thank from of depth our hearts.
We thankful to our Principle Dr. SRI VASTAV, Ph.D for permitting and
encouraging us in during this project.
We deeply intended to Dr. ARUL DALTON, Ph.D, and Head of the
Department, CSE whose motivation and constant encouragement has led to pursue a
project in the field of software development.
We very much obliged and thankful to our internal guide Ms. SANKU
VASAVI,MTech CSE for providing this opportunity and constant encouragement
given by her during the course. We grateful to her valuable guidance and suggestions
during our project work.
Our Parents have put ourself ahead of themselves. Because of their hard
work and dedication, we have had opportunity beyond our wildest dreams. Our heart
felt thanks to them for giving us all we ever needed to be successful student and
individual.
Finally we express our thanks to all our other professors, classmates,
friends, neighbours and our family members who helped us for the completion of our
project and without infinite love and patience this would never have been possible.

B.Uma devi 196M1A0572

V.Sruthi - 196M1A0593
N.Alekya - 196M1A05B5
A. Rupa bindu -196M1A05B6
B. Sravani -206M5AO502
I

CONTENT
ABSTRACT

LIST OF FIGURES AND TABLES

LIST OF SCREENS

1. INTRODUCTION 1-6

1.1 Motivation

1.2 Problem Defination

1.3 Objective Of Project

1.4 Limitations Of Project

1.5 Organization Of Documentation

2. LITERATURE SURVEY 7-13

2.1 Introduction

2.2 Existing System

2.3.1 Disadvantages Of Existing System

2.3 Proposed System

2.3.1 Advantages of proposed system

3. ANALYSIS 14-19

3.1Software Requirement Specification

3.1 Software Requirement

3.2 Hardware Requirement

3.2 Content Diagram Of Project

I
II

3.3 Algorithms And Flowcharts

3.3.1 Algorithms

3.3.2 flowchart

4. DESIGN 20-25

4.1 Uml Diagrams

4.1.1 Use Case Diagram

4.1.2 Class Diagram
4.1.3 Sequence Diagram
4.2 Data flow Diagram

4.3 Module Design And Organization

5. IMPLEMENTATION AND RERSULT 26-49

5.1 Introduction

5.2 Explanation Of Key Functions

5.3 Method Of Implementation

5.3.1 Source code

5.3.2 Output Screens

5.3.3 Result Analysis

6. TESTING AND VALIDATION 50-57

6.1 INTRODUCTION

6.2 DESIGN OF TEST CASES AND SCENARIOS7

II
III

7. VALIDAION 58-61

7.2 CONCLUSION

8. CONCLUSION 62

REFERENCES 63-64

III
IV

ABSTRACT

Cyberbullying (CB) has become increasingly prevalent in social media platforms.

With the popularity and widespread use of social media by individuals of all ages, it is
vital to make social media platforms safer from cyberbullying. This paper presents a
hybrid deep learning model, called DEA-RNN, to detect CB on Twitter social media
network. The proposed DEA-RNN model combines Elman type Recurrent Neural
Networks (RNN) with an optimized Dolphin Echolocation Algorithm (DEA) for
finetuning the Elman RNN’s parameters and reducing training time. We evaluated
DEA-RNN thoroughly utilizing a dataset of 10000 tweets and compared its
performance to those of state-of-the-art algorithms such as Bi-directional long short
term memory (Bi-LSTM), RNN, SVM, Multinomial Naive Bayes (MNB), Random
Forests (RF). The experimental results show that DEA-RNN was found to be superior
in all the scenarios. It outperformed the considered existing approaches in detecting
CB on Twitter platform. DEA-RNN was more efficient in scenario 3, where it has
achieved an average of 90.45% accuracy, 89.52% precision, 88.98% recall, 89.25%
F1-score, and 90.94% specificity. INDEX TERMS Cyber-bullying, tweet
classification, Dolphin Echolocation algorithm, Elman recurrent neural networks,
short text topic modeling, cyberbullying detection, social media.

IV
V

LIST OF FIGURES

S.No FIGURE NAME Page.No

Flow chart diagram 19

1

1 System Architecture 20

2 Use Case Diagram 21

4 Class Diagram 22

5 Sequence Diagram 23

6 Data Flow Diagram 24

V
VI

LIST- SCREENS

S. No SCREEN NAME Page No

1 Home Page 41
2 User Registration Form 41
3 User Login Form 41
4 User Home 42
5 Getting prediction 42
6 Trained and tested results 43
7 Admin login 43
8 Admin home page 43
9 View registered users 44
10 View trained and tested accuracy 44
11 All predicting corona virus ratio 45
12 Overall prediction ratio view 45

VI
VII

VI
I
1

1.INTRODUCTION

1.1 MOTIVATION

Social media networks such as Face book, Twitter, Flickr, and Instagram have
become the preferred online platforms for interaction and socialization among people
of all ages. While these platforms enable people to communicate and interact in
previously unthinkable ways, they have also led to malevolent activities such as
cyber-bullying. Cyber bullying is a type of psychological abuse with a significant
impact on society. Cyber-bullying events have been increasing mostly among young
people spending most of their time navigating between different social media
platforms. Particularly, social media networks such as Twitter and Face book are
prone to CB because of their popularity and the anonymity that the Internet provides
to abusers. In India, for example, 14 percent of all harassment occurs on Face book
and Twitter, with 37 percent of these incidents involving youngsters [1]. Moreover,
cyber bullying might lead to serious mental issues and adverse mental health effects.
Most suicides are due to the anxiety, depression, stress, and social and emotional
difficulties from cyber-bullying events [2]_[4]. This motivates the need for an
approach to identify cyber bullying in social media messages (e.g., posts, tweets, and
comments).
In this article, we mainly focus on the problem of cyber bullying
detection on the Twitter platform. As cyber bullying is becoming a prevalent problem
in Twitter, the detection of cyber bullying events from tweets and provisioning
preventive measures are the primary tasks in battling cyber bullying threats [5].
Therefore, there is a greater need to increase the research on social networks-based
CB in order to get greater insights and aid in the development of effective tools and
approaches to effectively combat cyber bullying problem [6]. Manually monitoring
and controlling cyber bullying on Twitter platform is virtually impossible [7].
Furthermore, mining social media messages for cyber bullying detection is quite
difficult. For example, Twitter messages are often brief, full of slang, and may include
emojis, and gifs, which makes it impossible to deduce individuals' intentions and
meanings purely from social media messages. Moreover, bullying can be difficult to
detect if the bully uses strategies like sarcasm or passive-aggressiveness to conceal it.

1
2

Despite the challenges that social media messages bring, cyber bullying detection on
social media is an open and active research topic. Cyber bullying detection within the
Twitter platform has largely been pursued through tweet classification and to a certain
extent with topic modeling approaches. Text classification based on supervised
machine learning (ML) models are commonly used for classifying tweets into
bullying and non-bullying tweets [8]_[17]. Deep learning (DL) based classifiers have
also been used for classifying tweets into bullying and non-bullying tweets [7],
[18]_[22]. Supervised classifiers have low performance in case the class labels are
unchangeable and are not relevant to the new events [23]. Also, it may be suitable
only for a pre-determined collection of events, but it cannot successfully handle
tweets that change on the _y. Topic modeling approaches have long been utilized as
the medium to extract the vital topics from a set of data to form the patterns or classes
in the complete dataset. Although the concept is similar, the general unsupervised
topic models cannot be efficient for short texts, and hence specialized unsupervised
short text topic models were employed [24]. These models effectively identify the
trending topics from tweets and extract them for further processing. These models
help in leveraging the bidirectional processing to extract meaningful topics. However,
these unsupervised models require extensive training to obtain sufficient prior
knowledge, which is not adequate in all cases [25]. Considering these limitations, an
efficient tweet classification approach must be developed to bridge the gap between
the classifier and the topic model so that the adaptability is significantly proficient.

In this article, we propose a hybrid deep learning-based approach,

called DEA-RNN, which automatically detects bullying from tweets. The DEA-RNN
approach combines Elman type Recurrent Neural Networks (RNN) with an improved
Dolphin Echolocation Algorithm (DEA) for _netuning the Elman RNN's parameters.
DEA-RNN can handle the dynamic nature of short texts and can cope with the topic
models for the effective extraction of trending topics. DEA-RNN outperformed the
considered existing approaches in detecting cyber bullying on the Twitter platform in
all scenarios and with various evaluation metrics.
The contributions of this article can be summarized as the following
_ Develop an improved optimization model of DEA for use to automatically tune the
RNN parameters to enhance the performance;

2
3

_ Propose DEA-RNN by combining the Elman type RNN and the improved DEA for
optimal classification of tweets;
_ A new Twitter dataset is collected based on cyber bullying keywords for evaluating
the performance of DEA-RNN and the existing methods; and
_ The efficiency of DEA-RNN in recognizing and classifying cyber bullying tweets is
assessed using Twitter datasets. The thorough experimental results reveal that DEA-
RNN outperforms other competing models in terms of recall, precision, accuracy, F1
score, and specificity.
The rest of this article is structured as the following: Recent related works
are reviewed and analyzed in Section II. The proposed DEA-RNN model is described
in Section III. Section IV discusses the experimental analysis, performance metrics,
and results analysis. The discussion is introduced
in Section V. Finally, Section VI offers the conclusion and possible future direction.

1.2 PROBLEM DEFINATION

1. The Cyber-attacks are growing faster than usual rate, it became evident that
necessary steps should be taken in-order to get them under control. Among various
cyber-attacks, Phishing websites is one of the popular and commonly used attack
to steal users personal information and financial information by manipulating the
website URL and IP addresses.

2. The main focus in this project is to implement the better model for detecting
these phishing websites using ML algorithms.

1.3 OBJECTIVE OF THE PROJECT

This paper presents a phishing URL dataset using legitimate login websites to obtain
the URLs from such pages. Then, we evaluate machine and deep learning techniques
for recommending the method with higher accuracy. Next, we show how models
trained with legitimate homepages struggle to classify legitimate login URLs,
demonstrating our hypothesis about phishing detection and legitimate login URLs.

3
4

Additionally, we show how the accuracy decrease with the time on models trained
with datasets from 2016 and evaluated on data collected in 2020. Finally, we provide
an overview of current phishing encounters, explaining attacker tricks and
approaches.

1.4 LIMITATIONS OF PROJECT

there are limitations and risks that need to be addressed to ensure success. The three

primary constraints that project managers should be familiar with are time, scope, and

cost. These are frequently referred to as the triple constraints or the project

management triangle.

Assumptions are things that we believe to be true and which we therefore build into

the project plan. Constraints are things that we know to be true and which must be

accounted for in the plan so that we can work around them. And risks are factors that

we are aware of but whose occurrence is uncertain.

 Stretched resources.

 Operational mishaps.

 Low performance.

 Lack of clarity.

 Scope creep.

 High costs.

 Time crunch.

4
5

1.6 ORGANIZATION OF THE REPORT

Chapter 1
Introduction:the overview provides the basic layout and the insight about the work
proposed. It briefs the entire need of the currently proposed work.

Problem definition: A problem statement is a concise description of an issue to be

addressed or a condition to be improved upon. We have identified the gap between
addressed or a condition to be improved upon. We have identified the gap between
the current existing work and the desired work. Considering the needs of society, we
have made an effort to approach the existing problems.

Objectives: A project objective describes the desired results of the work. We have
mentioned about the work we are trying to accomplish in this section.

limitations:

Chapter 2
Literature Survey: the purpose of a literature review is to gain an understanding of
the existing resources to a particular topic or area of study. We have referred to many
research papers relevant to our work in a better way.

Chapter 3
System Analysis: System Analysis is a document that describes about the existing
system and proposed system in the project. And also describes about advantages and
disadvantages in the project.

System Requirements and Specifications: System Requirements and Specifications

is a document that describes the nature of a project, software or application. This
section contains the brief knowledge about the functional and non – functional that
are needed to implement the project

Chapter 4
System design: System design is a document that describes about the project
modules, Activity diagram, Use Case Diagram, Data Flow Diagram, and Sequence
Diagram detailed in the project.

Chapter 5
Implementation: Implementation is a document that describes about the detailed
concepts of the project. Also describes about the algorithm with their detailed steps.
And also, about the codes for implementation of the algorithm

5
6

Result: The result of project describes the output of the project and it gives
certain output for certain input

Chapter 6
Testing: Testing is a document that describes about the a. Methods of testing: This
contains the information about Unit testing, Validation testing, Functional testing,
Integration testing, User Acceptance testing. b. Test Cases: In Test Cases we contain
the detailed description about program

Chapter 7
Conclusion: Conclusion and Future Enhancement is a document that describes about
the brief summary of the project and undetermined events that will occur in that time.

6
7

2.LITERATURE SURVEY
2.1 INTRODUCTION
This study extracted features using Word2Vec and TF-IDF methods. Dalvi et al. [12]

[27] used SVM and Random Forests (RF) models with TF-IDF for feature extraction

for detecting cyberbullying in tweets. Although SVM in these models achieved high

performance, the model complexity increases when the class labels are increased. Al-

garadi et al. [28] investigated cyberbullying identification using different ML

classifiers such as RF, Naïve Bayes (NB), and SVM based on various extracted

features from Twitter such as (tweet content, activity, network, and user). Huang et

al.

[29] suggested an approach for identifying CB from social media, which integrated

the social media features and textual content features. The features are ranked using

IG method. Well-known classifies such as NB, J48, and Bagging and Dagging are

utilized. The findings implied that social characteristics could aid in increasing the

accuracy of cyberbullying detection. Squicciarini et al. [30] utilized a decision tree

(C4.5) classifier with a social network, personal and textual features to identify

Cyberbullying and cyberbullying prediction on social networks like spring.me, and

MySpace. Balakrishnan et al. [31] utilized different ML algorithms such as RF, NB,

and J48 to detect cyberbullying events from tweets and classify tweets to different

cyberbullying classes such as aggressors, spammer, bully, and 25858 VOLUME 10,

2022 IEEE Access and Machine Learning,Volume:10,Issue Date:March.11,2022.

normal. The study concluded that the emotional feature does not impact the detection

rate. Despite its efficiency, this model is limited to a small dataset with fewer class

labels. Alam et al. [32] proposed an ensemble-based classification approach using the

single and double ensemble-based voting model. These ensemble-based voting

7
8

models utilized decision tree, LR, and Bagging ensemble model classifiers for the

classification while utilizing mutual information bigrams and unigram TF-IDF as

feature extraction models. On analysis over the Twitter dataset, the Bagging ensemble

model provided the best precision but considered other parameters. Although, these

ensemble models reduced the training and execution time for classification, the major

limitation comes when utilized sarcasm tweets and multiple-meaning acronym terms.

Chia et al. [8] also utilized different ML and feature engineering-based approaches to

classify irony and sarcasm from cyber-bullying tweets. In this approach, many

classifiers and feature selection methods were tested; while this approach greatly

detects the sarcasm and irony terms among cyber-bullying tweets, the detection rate is

still very low [33]. Similarly, Rafiq et al. [17] utilized decision tree, AdaBoost, NB,

and Randon Forest classifier to identify the instances of cyberbullying in a Vine

dataset. Authors collected the Vine media dataset and labeled it using Crowd-Sourced

and CrowdFlower websites. They utilized the comments, unigrams, media

information, and profile as the features. Nahar et al. [34] suggested a semi-supervised

learning method for detecting CB in social media in which training data samples are

augmented, and a fuzzy SVM method is applied. The augmented training approach

expands and extracts the training set from the unclassified streaming text

automatically. The learning is performed using a small limited training set given as an

initial input. The suggested method overcomes the dynamic and complex character of

streaming data. Xu et al. [35] provided many off-the-shelf methods, including LDA

and LSA-based modeling and Bagof-Words models for predicting bullying traces on

Twitter. A personalized cyberbullying detection framework, namely PI-Bully, was

introduced by Cheng et al. [36] to detect cyberbullying from the Twitter dataset. PI-

Bully composes three elements: a global element that determines the characteristics

8
9

that all users have in common, a personalized element that captures the distinctive

features of each user, and a peer influence element capable of quantifying the various

influences of other users people. Deep learning (DL) based approaches for

cyberbullying detection in tweets have also been proposed in the literature. N.

Yuvaraj et al. [9] used Artificial Neural Network (ANN) and Deep Reinforcement

Learning (DRL) to classify cyberbullying tweets. However, this approach has higher

computational complexity. Chen et al. [37] used a text classification model based on

CNN and 2-D TF-IDF features to enhance the sentiment analysis task performance.

The experimental results showed that the CNN model obtained optimal results

compared to the baselines LR and SVM models. Agrawal [16] utilized LSTM with

Transfer Learning for cyberbullying detection on several social media networks. A

new representation learning approach named smSDA (Semantic-Enhanced

Marginalized Denoising Autoencoder) was suggested by Zhao et al. [38] for detecting

cyberbullying. smSDA produced discriminative and robust representations.

Following that, The numerical representations that have been learned can be input

into SVM. Zhang [39] Suggested a new model which integrates the Gated Recurrent

unit Network GRU layers and CNN layers to detect hate speech. Al-Hassan and Al-

Dossari [19] utilized SVM as the baseline classifier and compared it against four DL

models, namely CNN + LTSM, LTSM, CNN + GRU, and GRU to detect

cyberbullying hate speech in Arabic tweets. However, the CNN+LSTM and

CNN+GRU complexity is higher and might not be effective in handling larger

datasets. Natarajan Yuvaraj et al. [18] proposed a new classification model for CB

detection from Twitter data. It used deep decision-tree classification with multi-

feature based AI for tweet classification. The deep decision tree classifier has been

designed by integrating the hidden layers of deep neural networks with the decision

9
10

tree classifier. This approach also utilized three feature selection approaches: Chi-

Square, Pearson Correlation, and IG. However, it cannot handle high-dimensional

data with such accuracy. Fang et al. [20] designed a classification model that

combines a selfAttention mechanism and bi-directional Gated Recurrent Unit (Bi-

GRU) to detect cyberbullying in tweets. This model employed merit for learning the

underlying relationships between words using BI-GRU and used it together with a

self-attention mechanism to improve the cyberbullying tweets classification process.

However, the context-independent behavior of the attention network creates

limitations in learning all relationships between the tweets. Pericherla and

Ilavarasan

[33] suggested a transformer network-based word embedding model to classify CB

tweets. This model utilizes Light Gradient Boosting Machine to classify the tweets

and RoBERTa to create word embedding. This approach overcomes the context-

independent limitations of traditional word embedding methods. Yet, this model has a

higher training time compared to the CNN models. Paul and Saha [40] proposed a

model for identifying cyberbullying, namely CyberBERT, based on the BERT.

Iwendi et al. [21] introduced a model to detect cyberbullying based on Bi-LSTM and

RNN. This model showed that the RNN could achieve high performance, but still, the

Bi-LSTM has significantly high efficiency. In some cases, CNN also performs better.

Akhter et al. [41] performed many DL models such as LSTM, CLSTM, CNN, and

BLSTM, and other ML models to discover an abusive language from Urdu social

media text. Some other studies utilized CNN’s to enhance the cyberbullying detection

[42]–[46]. Tripathy et al. [47] proposed a fine-tuning approach for detecting CB based

on ALBER. Agarwal et al. [7] utilized RNN based on Under-Sampling and Class

Weighting. These modifications helped the RNN model to perform better than the

LSTM model

1
11

2.2 EXISTING SYSTEM:

Purnamasari et al. [26] utilized the SVM and Information Gain(IG) based feature
selection method for detecting cyberbullying events in tweets. Muneer and

Fati [11] used various classifiers, namely AdaBoost(ADB), Light Gradient Boosting
Machine (LGBM), SVM, RF, Stochastic Gradient Descent (SGD), Logistic
Regression (LR), and MNB, and for cyberbullying events identification in tweets.
This study extracted features using Word2Vec and TF-IDF methods.

Dalvi et al. [12] [27] used SVM and Random Forests (RF) models with TF-IDF for
feature extraction for detecting cyberbullying in tweets. Although SVM in these
models achieved high performance, the model complexity increases when the class
labels are increased. Al-garadi et al. [28] investigated cyberbullying identification
using different ML classifiers such as RF, Naïve Bayes (NB), and SVM based on
various extracted features from Twitter such as (tweet content, activity, network, and
user). Huang et al. [29] suggested an approach for identifying CB from social media,
which integrated the social media features and textual content features.

The features are ranked using IG method. Well-known classifies such as NB, J48, and
Bagging and Dagging are utilized. The findings implied that social characteristics
could aid in increasing the accuracy of cyberbullying detection. Squicciarini et al.
[30] utilized a decision tree (C4.5) classifier with a social network,

personal and textual features to identify Cyberbullying and cyberbullying prediction

on social networks like spring.me, and MySpace. Balakrishnan et al. [31] utilized
different ML algorithms such as RF, NB, and J48 to detect cyberbullying events from
tweets and classify tweets to different cyberbullying classes such as aggressors,
spammer, bully, and normal. The study concluded that the emotional feature does

not impact the detection rate. Despite its efficiency, this model is limited to a small
dataset with fewer class labels.

1
12

Alam et al. [32] proposed an ensemble-based classification approach using the single
and double ensemble-based voting model. These ensemble-based voting models
utilized decision tree, LR, and Bagging ensemble model classifiers for the
classification while utilizing mutual information bigrams and unigram TF-IDF as
feature extraction models. On analysis over the Twitter dataset, the Bagging ensemble
model provided the best precision but considered other parameters. Although, these
ensemble models reduced the training and execution time for classification, the major
limitation comes when utilized sarcasm tweets and multiple-meaning acronym terms.
Chia et al. [8] also utilized different ML and feature

engineering-based approaches to classify irony and sarcasm from cyber-bullying

tweets. In this approach, many classifiers and feature selection methods were tested;
while this approach greatly detects the sarcasm and irony terms among cyber-bullying
tweets, the detection rate is still very low [33].

2.2.1 DISADVANTAGES OF EXITING SYSTEM

 The system is not implemented cyberbullying detection due to absence of an
effective ML classifiers.
 The system is not implemented DEA-RNN techniques which lead very less
prediction.

2.3 PROPOSED SYSTEM

In this article, we propose a hybrid deep learning-based approach, called DEA-
RNN, which automatically detects bullying from tweets. The DEA-RNN
approach combines Elman type Recurrent Neural Networks (RNN) with an
improved Dolphin Echolocation Algorithm (DEA) for fine tuning the Elman
RNN's parameters. DEA-RNN can handle the dynamic nature of short texts and
can cope with the topic models for the effective extraction of trending topics.
DEA-RNN outperformed the considered existing approaches in detecting
cyberbullying on the Twitter platform in all scenarios and with various
evaluation metrics. The contributions of this article can be summarized as the
following:
_ Develop an improved optimization model of DEA for use to automatically
tune the RNN parameters to enhance the performance;

1
13

_ Propose DEA-RNN by combining the Elman type RNN and the improved
DEA for optimal classification of tweets;
_ A new Twitter dataset is collected based on cyberbullying keywords for
evaluating the performance of DEA-RNN and the existing methods; and
_ The efficiency of DEA-RNN in recognizing and classifying cyberbullying
tweets is assessed using Twitter datasets. The thorough experimental results
reveal that DEA-RNN outperforms other competing models in terms of recall,
precision, accuracy, F1 score, and specificity
2.3.1 ADVANTAGES OF PROPOSED SYSTEM

The proposed system effectively identifies the trending topics from tweets and
extracts them for further processing. An effective models help in leveraging the
bidirectional processing to extract meaningful topics.

An effective system which is mainly tested and trained by SVM, Multinomial

Naive Bayes (MNB),Random Forests (RF) classify

1
14

3.ANALYSIS

3.1 SOFTWARE REQUIREMENT SPECIFICATION

System Requirement Specifications is used for the programming contexture that are

basically being for the functionality of the system can do, and also for the work

behind the organization for describing and to understand the client’s needs. The

purpose of SRS gives the requirement to be master planned of a system or sub-

system. It generally provides for the capable clients needful conditions at a particular

instance of time before the work is finalize. A software requirements specification is a

complete that describes the behavior of system to be developed. Use case techniques

can be used to find the product of functional requirements or supplementary

requirements. A non-functional requirement performs the engineering requirements,

quality of standards

3.1.1 HARDWARE REQUIREMENTS:

➢ Processor - Pentium –IV

➢ RAM - 4 GB (min)

➢ Hard Disk - 20 GB

➢ Key Board - Standard Windows Keyboard

➢ Mouse - Two or Three Button Mouse

➢ Monitor - SVGA

1
15

3.1.2 SOFTWARE REQUIREMENTS:

4 Operating system : Windows 7 Ultimate.

5 Coding Language : Python.

6 Front-End : Python.

7 Back-End : Django-ORM

8 Designing : Html, css, javascript.

9 Data Base : MySQL (WAMP Server)

3.2 CONTENT DIAGRAM OF PROJECT

The Content diagram is an extension of UML notation. The purpose of the Content
diagram is to generate or represent a project structure (diagrams) and relations
between them. The Content table works as a table of contents for a project.

1
16

3.3ALGORITHM AND FLOWCHART

3.3.1 ALGORITHMS

Decision tree

Decision tree classifiers are used successfully in many diverse areas. Their most
important feature is the capability of capturing descriptive decision making
knowledge from the supplied data. Decision tree can be generated from training sets.
The procedure for such generation based on the set of objects (S), each belonging to
one of the classes C1, C2, …, Ck is as follows:\

Step 1. If all the objects in S belong to the same class, for example Ci, the decision
tree for S consists of a leaf labeled with this class

Step 2. Otherwise, let T be some test with possible outcomes O1, O2,…, On. Each
object in S has one outcome for T so the test partitions S into subsets S1, S2,… Sn
where each object in Si has outcome Oi for T. T becomes the root of the decision tree
and for each outcome Oi we build a subsidiary decision tree by invoking the same
procedure recursively on the set Si.

Gradient boosting

Gradient boosting is a machine learning technique used in regression and

classification tasks, among others. It gives a prediction model in the form of an
ensemble of weak prediction models, which are typically decision trees.[1][2] When a
decision tree is the weak learner, the resulting algorithm is called gradient-boosted
trees; it usually outperforms random forest.A gradient-boosted trees model is built in
a stage-wise fashion as in other boosting methods, but it generalizes the other methods
by allowing optimization of an arbitrary differentiable loss function.

K-Nearest Neighbors (KNN)

 Simple, but a very powerful classification algorithm

 Classifies based on a similarity measure
 Non-parametric
 Lazy learning

1
17

Does not “learn” until the test example is given

Whenever we have a new data to classify, we find its K-nearest neighbors from the
training data

Logistic regression Classifiers

Logistic regression analysis studies the association between a categorical dependent

variable and a set of independent (explanatory) variables. The name logistic
regression is used when the dependent variable has only two values, such as 0 and 1
or Yes and No. The name multinomial logistic regression is usually reserved for the
case when the dependent variable has three or more unique values, such as Married,
Single, Divorced, or Widowed. Although the type of data used for the dependent
variable is different from that of multiple regression, the practical use of the procedure
is similar.

Naïve Bayes

The naive bayes approach is a supervised learning method which is based on a

simplistic hypothesis: it assumes that the presence (or absence) of a particular feature
of a class is unrelated to the presence (or absence) of any other feature .

Yet, despite this, it appears robust and efficient. Its performance is comparable to
other supervised learning techniques. Various reasons have been advanced in the
literature. In this tutorial, we highlight an explanation based on the representation
bias. The naive bayes classifier is a linear classifier, as well as linear discriminant
analysis, logistic regression or linear SVM (support vector machine). The difference
lies on the method of estimating the parameters of the classifier (the learning bias)

Random Forest

Random forests or random decision forests are an ensemble learning method

for classification, regression and other tasks that operates by constructing a multitude
of decision trees at training time. For classification tasks, the output of the random
forest is the class selected by most trees. For regression tasks, the mean or average
prediction of the individual trees is returned. Random decision forests correct for
decision trees' habit of overfitting to their training set. Random forests generally

1
18

outperform decision trees, but their accuracy is lower than gradient boosted trees.
However, data characteristics can affect their performance.

Random forests are frequently used as "blackbox" models in businesses, as they

generate reasonable predictions across a wide range of data while requiring little
configuration.

SVM

In classification tasks a discriminant machine learning technique aims at

finding, based on an independent and identically distributed (iid) training dataset, a
discriminant function that can correctly predict labels for newly acquired instances.
Unlike generative machine learning approaches, which require computations of
conditional probability distributions, a discriminant classification function takes a data
point x and assigns it to one of the different classes that are a part of the classification
task. Less powerful than generative approaches, which are mostly used when
prediction involves outlier detection, discriminant approaches require fewer
computational resources and less training data, especially for a multidimensional
feature space and when only posterior probabilities are needed. From a geometric
perspective, learning a classifier is equivalent to finding the equation for a
multidimensional surface that best separates the different classes in the feature space

3.3.1 FLOWCHART

A flowchart is a type of diagram that represents an algorithm, workflow or

process. The flowchart shows the steps as boxes of various kinds, and their
order by connecting the boxes with arrows. This diagrammatic representation
illustrates a solution model to a given problem.

1
19

Service provider:

Remote user:

1
20

4. ANALYSIS

4.1 UML DIAGRAMS

UML stands for Unified Modeling Language. UML is a standardized general-purpose

modeling language in the field of object-oriented software engineering. The standard
is managed, and was created by, the Object Management Group.
The goal is for UML to become a common language for creating models of object
oriented computer software. In its current form UML is comprised of two major
components: a Meta-model and a notation. In the future, some form of method or
process may also be added to; or associated with, UML.The Unified Modeling
Language is a standard language for specifying, Visualization, Constructing and
documenting the artifacts of software system, as well as for business modeling and
other non-software systems.
The UML represents a collection of best engineering practices that have proven
successful in the modeling of large and complex systems.
The UML is a very important part of developing objects oriented software and the
software development process. The UML uses mostly graphical notations to express
the design of software projects.

GOALS:
The Primary goals in the design of the UML are as follows:
1. Provide users a ready-to-use, expressive visual modeling Language so that
they can develop and exchange meaningful models.
2. Provide extendibility and specialization mechanisms to extend the core
concepts.
3. Be independent of particular programming languages and development
process.
4. Provide a formal basis for understanding the modeling language.
5. Encourage the growth of OO tools market.
6. Support higher level development concepts such as collaborations,
frameworks, patterns and components.
7. Integrate best practices

2
21

4.1.1 USE CASE DIAGRAM:

A use case diagram in the Unified Modeling Language (UML) is a type of

behavioral diagram defined by and created from a Use-case analysis. Its purpose is to
present a graphical overview of the functionality provided by a system in terms of
actors, their goals (represented as use cases), and any dependencies between those
use cases. The main purpose of a use case diagram is to show what system functions
are performed for which actor.

2
22

4.1.2 CLASS DIAGRAM:

In software engineering, a class diagram in the Unified
Modeling Language (UML) is a type of static structure diagram that describes the
structure of a system by showing the system's classes, their attributes, operations (or
methods), and the relationships among the classes. It explains which class contains
information.

2
23

4.1.3 SEQUENCE DIAGRAM:

A sequence diagram in Unified Modeling Language (UML) is a kind of
interaction diagram that shows how processes operate with one another and in what
order. It is a construct of a Message Sequence Chart. Sequence diagrams are
sometimes called event diagrams, event scenarios, and timing diagrams

4.2 DATA FLOW DIAGRAM:

2
24

1. The DFD is also called as bubble chart. It is a simple graphical formalism that
can be used to represent a system in terms of input data to the system, various
processing carried out on this data, and the output data is generated by this
system.
2. The data flow diagram (DFD) is one of the most important modeling tools. It
is used to model the system components. These components are the system
process, the data used by the process, an external entity that interacts with the
system and the information flows in the system.
3. DFD shows how the information moves through the system and how it is
modified by a series of transformations. It is a graphical technique that depicts
information flow and the transformations that are applied as data moves from
input to output.
4. DFD is also known as bubble chart. A DFD may be used to represent a system
at any level of abstraction. DFD may be partitioned into levels that represent
increasing information flow and functional detail.

2
25

4.3 MODULE DESGIN AND ORGANIZATION

 Service provider
 View and authorize users
 Remote user

MODULES DESCRIPTION:

Service Provider

In this module, the Service Provider has to login by using valid user name and
password. After login successful he can do some operations such as
Login, Train & Test URL Data Sets, View URL Data Sets Trained and Tested
Accuracy in Bar Chart, View URL Data Sets Trained and Tested Accuracy Results,
View Prediction Of URL Type, View URL Type Ratio, Download Predicted Data
Sets, View URL Type Ratio Results, View All Remote Users.

View and Authorize Users

In this module, the admin can view the list of users who all registered. In this, the
admin can view the user’s details such as, user name, email, address and admin
authorizes the users.

Remote User

In this module, there are n numbers of users are present. User should register before
doing any operations. Once user registers, their details will be stored to the database.
After registration successful, he has to login by using authorized user name and
password. Once Login is successful user will do some operations like REGISTER
AND LOGIN, PREDICT URL TYPE, VIEW YOUR PROFILE..

2
26

5.IMPLEMENTATION AND RESULT

5.1 INTRODUCTION
(i) data collection, (ii) data annotation, (iii) pre-processing and data cleansing, (iv)
feature extraction and feature selection, and (v) classification.

A. DATA COLLECTION
The input dataset is made up of tweets collected through Twitter API streaming with
the help of around 32 cyber-bullying keywords. Idiot, ni**er, LGBTQ (le***an, g*y,
bisexual, transgender, and queer), whore, pussy, faggot, shit, sucker, slut, donkey,
live, afraid, moron, poser, rape, fuck, fucking, ugly, bitch, ass, whale, etc. are some of
the keywords as recommended in psychology literature [30], [36], [57]. Whereas the
other keywords such as ban, kill, die, evil, hate, attack, terrorist, threat, racism, black,
Muslim, Islam, and Islamic were suggested in [39]

B. DATA ANNOTATION
This section mainly concentrates on annotating and labeling the selected tweets from
the original Twitter dataset. After selecting 10000 tweets randomly from the
collecting tweets, the selected tweets were labeled manually into two labels, either
‘‘0’’ non-cyber bullying or ‘‘1’’ cyberbullying, by a set of three human annotators
over a period of one and half months. In the labeling procedure, the human annotators
labeled the instances based on whether it was considered to involve cyberbullying and
also the guidelines described in detail in [57]. The making decision of the
cyberbullying instances depends on the following guidelines: character attacks,
insults, competence attacks, malediction, verbal abuse, teasing, name-calling,
mockery, threats, and physical appearance. Initially, each tweet was classified by two
annotators, and the level of agreement rate between the two annotators was 91%
approximately at this phase. Then, a third annotator was tasked with resolving the
discrepancies discovered during the initial annotation process. Finally, we obtained
the final dataset after resolving discrepancies and cleaned up the data, which
contained 10000 labeled tweets, among which 6,508 (0.65%) are non-cyberbullying,
and 3492 (0.35%) are cyberbullying tweets. By observing the number of
cyberbullying and non-cyberbullying tweets, the labeled Twitter dataset is
imbalanced. The number of tweets in classes is greatly variable

2
27

C. PRE-PROCESSING AND DATA CLEANSING

The data cleansing and pre-processing phase contain three sub-phases [58]. This
process is performed on the raw tweet dataset to form the finalized data as described
in the previous dataset. In the first sub-phase, noise removal such as URL removal,
hashtag/mentions removal, punctuation/symbol removal, and emoticon transformation
processes are performed. In the second sub-phase, Out of Vocabulary Cleansing such
as spell checking, acronym expansion, slang modification, elongated (repeated
Characters removal) are performed. In the final sub-phase, tweet transformations such
as lower-case conversion, stemming, word segmentation (tokenization), and stop
word filtering are conducted. These subphases are performed to enhance the tweets
and improve feature extraction and classification accuracy. Figure 2 shows the pre-
processing and data cleansing steps.

D. FEATURE EXTRACTION AND SELECTION

The features from the Twitter dataset are extracted using NLP tools such as
Word2Vec and TF-IDF, with the nouns, pronouns, and adjectives are considered as
primary feature contents, whereas the adverbs and verbs provide additional
information. Furthermore, the extraction of Part-of-Speech (POS) tags, function
words, and content word features can improve the classification performance .

E. DEA-RNN CLASSIFIER MODEL

DEA mimics the behaviors and the capability of dolphins to generate a kind of echo
(click sounds) during the hunting process [61]. Initially, the dolphin’s population is
initialized, and the search space alternatives for each feature are ordered in ascending
or descending order. For variable j, feature vectors Aj with the length LAj is
constructed, which includes all potential alternatives for the jth variable

5.2 EXPLANATION OF KEY FUNCTIONS

Phishers use login forms to retrieve and steal users’ data. As far as we are concerned,
the legitimate class in most phishing datasets are represented by URLs from their
homepages [18], [19]. However, most websites have their login form in different
locations, making models trained with such public datasets to be biased since the
URLs of homepages tend to be shorter and simpler than others. An example of this is

2
28

depicted in Figure 2. In this paper, we present an extended version of the Phishing

Index Login URL (PILU-60K) dataset [20] and we name it PILU-90K. PILU-90K
contains 90K URLs divided into three classes (see Figure 2): 30K legitimate URLs of
homepages, 30K legitimate login URLs and 30K phishing.

DATASETS
To test the model robustness against URLs collected in different periods, we used the five
phishing datasets shown in Table 3. These datasets are grouped into two different categories
depending on their recollection strategy: (i) category A: PWD2016, 1M-PD and PIU-60K
collected legitimate samples by inspecting the top-visited domains and (ii) category B:
Ebbu2017 and PLU-60K visited those websites and performed further actions: in the case of
Ebbu2017, its authors retrieved the inner URLs and, in the case of PLU60K, we looked for
the login form page. Therefore most of the URLs include a path. Table 4 shows the
distribution of sample structure within the datasets.

EXPERIMENTAL SETTINGS

Experiments are executed on an Intel Core i3 9100F at 3.6Ghz and 16 GB of DDR4

RAM. We used scikit-learn11 and Python 3 for the implementation of the different
experiments. For the machine learning experiments, we empirically assign the
parameters that returned the best accuracy on the three different phishing datasets.
These parameters are shown in Table 5. 11https://scikit-learn.org/stable/ We used the
averaged values of 10-fold cross-validation, reporting the accuracy (Eq. (3)), the F1-
Score (Eq. (4)), the precision (Eq. (1)) and the recall (Eq. (2)).

Precision = TP TP + FP (1)

Recall = TP TP + FN (2)

Accuracy = TP + TN TP + TN + FN + FP (3)

F1 = 2 · Precision · Recall Precision + Recall (4)

TP denotes the true positives, i.e., how many phishing websites were correctly
classified. FP refers to the false positives and represents the number of legitimate
samples wrongly classified as phishing. TN (i.e., the true negatives) denotes the
number of legitimate samples correctly classified. Finally, FN represents the false

2
29

negatives that represent the number of phishing websites misclassified as legitimate

ones. Regarding the clustering experiments, we used the same approach of Al-Nabki
et al. [50] for text representation, as explained in Section IV-A and, for the clustering,
we used the Agglomerative Hierarchical Clustering (AHC) [52]. The clustering
process is repeated four times, and each time we initialized the AHC with the number
n of the desired clusters, i.e. n ∈ {4, 5, 6, 7}.

5.3 METHOD OF IMPLEMENTATION

5.3.1 SOURCE CODE

-- phpMyAdmin SQL Dump

-- version 4.0.4
-- http://www.phpmyadmin.net
--
-- Host: localhost
-- Generation Time: Aug 19, 2022 at 01:12 PM
-- Server version: 5.6.12-log
-- PHP Version: 5.4.16

SET SQL_MODE = "NO_AUTO_VALUE_ON_ZERO";

SET time_zone = "+00:00";

/*!40101 SET @OLD_CHARACTER_SET_CLIENT=@@CHARACTER_SET_CLIENT

*/;
/*!40101 SET
@OLD_CHARACTER_SET_RESULTS=@@CHARACTER_SET_RESULTS */;
/*!40101 SET @OLD_COLLATION_CONNECTION=@@COLLATION_CONNECTION
*/;
/*!40101 SET NAMES utf8 */;

--
-- Database: `dearnn`
--
CREATE DATABASE IF NOT EXISTS `dearnn` DEFAULT CHARACTER SET
latin1 COLLATE latin1_swedish_ci;
USE `dearnn`;

--
-- Table structure for table `auth_group`
--

CREATE TABLE IF NOT EXISTS `auth_group` (

`id` int(11) NOT NULL AUTO_INCREMENT,
`name` varchar(80) NOT NULL,
PRIMARY KEY (`id`),
UNIQUE KEY `name` (`name`)
) ENGINE=InnoDB DEFAULT CHARSET=latin1 AUTO_INCREMENT=1 ;

2
30

--
-- Table structure for table `auth_group_permissions`
--

CREATE TABLE IF NOT EXISTS `auth_group_permissions` (

`id` int(11) NOT NULL AUTO_INCREMENT,
`group_id` int(11) NOT NULL,
`permission_id` int(11) NOT NULL,
PRIMARY KEY (`id`),
UNIQUE KEY
`auth_group_permissions_group_id_permission_id_0cd325b0_uniq`
(`group_id`,`permission_id`),
KEY
`auth_group_permissio_permission_id_84c5c92e_fk_auth_perm`
(`permission_id`)
) ENGINE=InnoDB DEFAULT CHARSET=latin1 AUTO_INCREMENT=1 ;

--
-- Table structure for table `auth_permission`
--

CREATE TABLE IF NOT EXISTS `auth_permission` (

`id` int(11) NOT NULL AUTO_INCREMENT,
`name` varchar(255) NOT NULL,
`content_type_id` int(11) NOT NULL,
`codename` varchar(100) NOT NULL,
PRIMARY KEY (`id`),
UNIQUE KEY
`auth_permission_content_type_id_codename_01ab375a_uniq`
(`content_type_id`,`codename`)
) ENGINE=InnoDB DEFAULT CHARSET=latin1 AUTO_INCREMENT=28 ;

--
-- Dumping data for table `auth_permission`
--

INSERT INTO `auth_permission` (`id`, `name`,

`content_type_id`, `codename`) VALUES
(1, 'Can add log entry', 1, 'add_logentry'),
(2, 'Can change log entry', 1, 'change_logentry'),
(3, 'Can delete log entry', 1, 'delete_logentry'),
(4, 'Can add permission', 2, 'add_permission'),
(5, 'Can change permission', 2, 'change_permission'),
(6, 'Can delete permission', 2, 'delete_permission'),
(7, 'Can add group', 3, 'add_group'),
(8, 'Can change group', 3, 'change_group'),
(9, 'Can delete group', 3, 'delete_group'),
(10, 'Can add user', 4, 'add_user'),
(11, 'Can change user', 4, 'change_user'),
(12, 'Can delete user', 4, 'delete_user'),
(13, 'Can add content type', 5, 'add_contenttype'),

3
31

(14, 'Can change content type', 5, 'change_contenttype'),

(15, 'Can delete content type', 5, 'delete_contenttype'),
(16, 'Can add session', 6, 'add_session'),
(17, 'Can change session', 6, 'change_session'),
(18, 'Can delete session', 6, 'delete_session'),
(19, 'Can add client register_ model', 7,
'add_clientregister_model'),
(20, 'Can change client register_ model', 7,
'change_clientregister_model'),
(21, 'Can delete client register_ model', 7,
'delete_clientregister_model'),
(22, 'Can add client posts_ model', 8,
'add_clientposts_model'),
(23, 'Can change client posts_ model', 8,
'change_clientposts_model'),
(24, 'Can delete client posts_ model', 8,
'delete_clientposts_model'),
(25, 'Can add feedbacks_ model', 9, 'add_feedbacks_model'),
(26, 'Can change feedbacks_ model', 9,
'change_feedbacks_model'),
(27, 'Can delete feedbacks_ model', 9,
'delete_feedbacks_model');

--
-- Table structure for table `auth_user`
--

CREATE TABLE IF NOT EXISTS `auth_user` (

`id` int(11) NOT NULL AUTO_INCREMENT,
`password` varchar(128) NOT NULL,
`last_login` datetime(6) DEFAULT NULL,
`is_superuser` tinyint(1) NOT NULL,
`username` varchar(150) NOT NULL,
`first_name` varchar(30) NOT NULL,
`last_name` varchar(150) NOT NULL,
`email` varchar(254) NOT NULL,
`is_staff` tinyint(1) NOT NULL,
`is_active` tinyint(1) NOT NULL,
`date_joined` datetime(6) NOT NULL,
PRIMARY KEY (`id`),
UNIQUE KEY `username` (`username`)
) ENGINE=InnoDB DEFAULT CHARSET=latin1 AUTO_INCREMENT=1 ;

--
-- Table structure for table `auth_user_groups`
--

CREATE TABLE IF NOT EXISTS `auth_user_groups` (

`id` int(11) NOT NULL AUTO_INCREMENT,
`user_id` int(11) NOT NULL,
`group_id` int(11) NOT NULL,
PRIMARY KEY (`id`),

3
32

UNIQUE KEY `auth_user_groups_user_id_group_id_94350c0c_uniq`

(`user_id`,`group_id`),
KEY `auth_user_groups_group_id_97559544_fk_auth_group_id`
(`group_id`)
) ENGINE=InnoDB DEFAULT CHARSET=latin1 AUTO_INCREMENT=1 ;

--
-- Table structure for table `auth_user_user_permissions`
--

CREATE TABLE IF NOT EXISTS `auth_user_user_permissions` (

`id` int(11) NOT NULL AUTO_INCREMENT,
`user_id` int(11) NOT NULL,
`permission_id` int(11) NOT NULL,
PRIMARY KEY (`id`),
UNIQUE KEY
`auth_user_user_permissions_user_id_permission_id_14a6b632_uni
q` (`user_id`,`permission_id`),
KEY
`auth_user_user_permi_permission_id_1fbb5f2c_fk_auth_perm`
(`permission_id`)
) ENGINE=InnoDB DEFAULT CHARSET=latin1 AUTO_INCREMENT=1 ;

--
-- Table structure for table `django_admin_log`
--

CREATE TABLE IF NOT EXISTS `django_admin_log` (

`id` int(11) NOT NULL AUTO_INCREMENT,
`action_time` datetime(6) NOT NULL,
`object_id` longtext,
`object_repr` varchar(200) NOT NULL,
`action_flag` smallint(5) unsigned NOT NULL,
`change_message` longtext NOT NULL,
`content_type_id` int(11) DEFAULT NULL,
`user_id` int(11) NOT NULL,
PRIMARY KEY (`id`),
KEY `django_admin_log_content_type_id_c4bce8eb_fk_django_co`
(`content_type_id`),
KEY `django_admin_log_user_id_c564eba6_fk_auth_user_id`
(`user_id`)
) ENGINE=InnoDB DEFAULT CHARSET=latin1 AUTO_INCREMENT=1 ;

--
-- Table structure for table `django_content_type`
--

CREATE TABLE IF NOT EXISTS `django_content_type` (

`id` int(11) NOT NULL AUTO_INCREMENT,
`app_label` varchar(100) NOT NULL,

3
33

`model` varchar(100) NOT NULL,

PRIMARY KEY (`id`),
UNIQUE KEY
`django_content_type_app_label_model_76bd3d3b_uniq`
(`app_label`,`model`)
) ENGINE=InnoDB DEFAULT CHARSET=latin1 AUTO_INCREMENT=10 ;

--
-- Dumping data for table `django_content_type`
--

INSERT INTO `django_content_type` (`id`, `app_label`, `model`)

VALUES
(1, 'admin', 'logentry'),
(3, 'auth', 'group'),
(2, 'auth', 'permission'),
(4, 'auth', 'user'),
(8, 'Client_Site', 'clientposts_model'),
(7, 'Client_Site', 'clientregister_model'),
(9, 'Client_Site', 'feedbacks_model'),
(5, 'contenttypes', 'contenttype'),
(6, 'sessions', 'session');

--
-- Table structure for table `django_migrations`
--

CREATE TABLE IF NOT EXISTS `django_migrations` (

`id` int(11) NOT NULL AUTO_INCREMENT,
`app` varchar(255) NOT NULL,
`name` varchar(255) NOT NULL,
`applied` datetime(6) NOT NULL,
PRIMARY KEY (`id`)
) ENGINE=InnoDB DEFAULT CHARSET=latin1 AUTO_INCREMENT=22 ;

--
-- Dumping data for table `django_migrations`
--

INSERT INTO `django_migrations` (`id`, `app`, `name`,

`applied`) VALUES
(1, 'Remote_User', '0001_initial', '2019-04-23
07:01:48.050781'),
(2, 'contenttypes', '0001_initial', '2019-04-23
07:01:49.494140'),
(3, 'auth', '0001_initial', '2019-04-23 07:02:03.837890'),
(4, 'admin', '0001_initial', '2019-04-23 07:02:05.832031'),
(5, 'admin', '0002_logentry_remove_auto_add', '2019-04-23
07:02:05.863281'),
(6, 'contenttypes', '0002_remove_content_type_name', '2019-04-
23 07:02:07.041015'),
(7, 'auth', '0002_alter_permission_name_max_length', '2019-04-
23 07:02:07.839843'),

3
34

(8, 'auth', '0003_alter_user_email_max_length', '2019-04-23

07:02:08.330078'),
(9, 'auth', '0004_alter_user_username_opts', '2019-04-23
07:02:08.361328'),
(10, 'auth', '0005_alter_user_last_login_null', '2019-04-23
07:02:08.921875'),
(11, 'auth', '0006_require_contenttypes_0002', '2019-04-23
07:02:08.953125'),
(12, 'auth', '0007_alter_validators_add_error_messages',
'2019-04-23 07:02:08.989257'),
(13, 'auth', '0008_alter_user_username_max_length', '2019-04-
23 07:02:09.785156'),
(14, 'auth', '0009_alter_user_last_name_max_length', '2019-04-
23 07:02:10.580078'),
(15, 'sessions', '0001_initial', '2019-04-23
07:02:11.764648'),
(16, 'Remote_User', '0002_clientposts_model', '2019-04-25
05:53:57.702132'),
(17, 'Remote_User', '0003_clientposts_model_usefulcounts',
'2019-04-25 10:00:02.521468'),
(18, 'Remote_User', '0004_auto_20190429_1027', '2019-04-29
04:57:32.672296'),
(19, 'Remote_User', '0005_clientposts_model_dislikes', '2019-
04-29 05:15:16.668390'),
(20, 'Remote_User', '0006_Review_model', '2019-04-29
05:19:26.382257'),
(21, 'Remote_User', '0007_clientposts_model_names', '2019-04-
30 04:45:46.472656');

--
-- Table structure for table `django_session`
--

CREATE TABLE IF NOT EXISTS `django_session` (

`session_key` varchar(40) NOT NULL,
`session_data` longtext NOT NULL,
`expire_date` datetime(6) NOT NULL,
PRIMARY KEY (`session_key`),
KEY `django_session_expire_date_a5c62663` (`expire_date`)
) ENGINE=InnoDB DEFAULT CHARSET=latin1;

--
-- Dumping data for table `django_session`
--

INSERT INTO `django_session` (`session_key`, `session_data`,

`expire_date`) VALUES
('08snhdsayx8ns06yhe19rdyvnabjn8iy',
'eyJ1c2VyaWQiOjF9:1oFx6B:3_h3tiP91115hfoAJl2N8umQpJF5Lg0b3BH7V
Qg2w4o', '2022-08-08 12:22:11.245658'),
('0doxd89lelsb62hh61y1u9xvhgis91ks',
'eyJ1c2VyaWQiOjE5fQ:1mDlBs:xxj-bSmT5wd80FZ0QPs9yzR1wUnf-
8naNjRRmgbWt-4', '2021-08-25 10:10:28.069873'),

3
35

('0jpcgnd1gmwbp3e8tw54e6nxjylsogyo',
'YmM4NjE0MDQ2MzBmYWIxNzIzNTkxZjBiN2I5M2MxMzQyYTE0YmMxODp7InVzZ
XJpZCI6Mn0=', '2020-02-21 08:52:28.687500'),
('1avxwyhltuaclq2zfq40bjmwzxgup6hp',
'eyJ1c2VyaWQiOjJ9:1ml5JO:9_XIwCjkeG7Vu6-k169B1gbde6aRX-
mqiwhgCKNqrRA', '2021-11-25 08:19:58.276708'),
('49qo7iki5uxczhyymi8ka7dnh6a2wva5',
'MmE4N2EzZmM3NTI1ODc3MjUxYjUxNWM3OWM4ZGExNWViMzRkN2MzYTp7Im5hb
WUiOjF9', '2019-05-08 09:19:45.486328'),
('4df7s82pddaszour6twx23d86058ppjq',
'ZmNkODA5MmI1ZGQ0Yjk5MmZlNzEyNTcwNTcxNjk2ZWYxZTE3NThkMjp7InVzZ
XJpZCI6NX0=', '2020-11-23 11:49:21.396484'),
('4io28d085qjfib7a5s2qbhc8qp4wfiva',
'eyJ1c2VyaWQiOjE2fQ:1mAtmi:oIUbcN3WzJiaWnxMBZ6eIGMTo8NS2y701Jl
pwqvzBUk', '2021-08-17 12:44:40.453750'),
('4x6b78w9rfcn34v650kd2j7oij6atr8p',
'Zjk0Y2RlYjc4OTJhNWMyZjQyNmM4ZGRhYTVjNmVlNDFhZGE4ZmU3NTp7InVzZ
XJpZCI6Nn0=', '2019-12-27 12:07:42.082031'),
('6h7f2lif2b5xzckoqy0dzpuf2ssln6u6',
'eyJ1c2VyaWQiOjJ9:1oOxNy:rn921d2FdMCS42DJKaBJnAlT20byTRiY2P4X7
Wm76Qc', '2022-09-02 08:29:46.357161'),
('7ai80sml3hwq0jsvdfl1vm5m13x8d8k9',
'eyJ1c2VyaWQiOjF9:1mt5vR:3pScFzSqVCPEshZhoFMwZb-
rfCX09_pAUFEuMV8fNKA', '2021-12-17 10:36:21.166280'),
('7ixdamflp4fqyjecn17bd7xfbsi7eowq',
'eyJ1c2VyaWQiOjEwfQ:1mBzQr:5DfHs08xtygiklJxfW3kZFCrxnrA_igxR5g
bDcKt2e8', '2021-08-20 12:58:37.947486'),
('7ph664obz14m207786d3oubrzgjnisom',
'eyJ1c2VyaWQiOjN9:1ml5U2:RJ7eMbREY4fk71sBmCItxM6E3kDDc-R-clIp-
QGiiuI', '2021-11-25 08:30:58.769500'),
('9vom7mmn5muyoiy8nytc9mxown1q1g2b',
'eyJ1c2VyaWQiOjE5fQ:1mDRHr:QCoJ_gmMMx_cxknA5j_5NlcTLnENHFouosR
uxYZlYbI', '2021-08-24 12:55:19.038051'),
('au3tqhab9csr4r2g5p8wxgktebzxone0',
'eyJ1c2VyaWQiOjExfQ:1mexd7:BmPTZn93Z2602ApV03LTh7BmDypyoNMN2YR
KctHrGF8', '2021-11-08 10:55:01.820399'),
('b9cu6cjsfqfm5mame5dy1ikpiiy7yn3w',
'OTk3NTk2YTE0NjM5MWQ0OGQ0MjY3NzBjNzdhOTc0ZWJhM2ZkMzdkMjp7InVzZ
XJpZCI6MX0=', '2019-05-09 11:00:08.480453'),
('bhfid9lacfwlfi5yu3rgdg1uo5fp2bq8',
'eyJ1c2VyaWQiOjE4fQ:1mBH4F:2wUorkPET_MGY07bWd-Zp-
9HZUsjS3bGCHCu1j6BN-s', '2021-08-18 13:36:19.192796'),
('ct13q5fpn94zvnij8ekixwzcky2imc5e',
'YWUzM2IzMWJiYmQ3YmY2YzlkMGFlNTM1YmU5ZGM4YjQ0MmY1YTc0NTp7InVzZ
XJpZCI6NH0=', '2019-05-14 11:44:10.978515'),
('e07j4duysh402dedtomm8icctvs9ljgy',
'MmE4N2EzZmM3NTI1ODc3MjUxYjUxNWM3OWM4ZGExNWViMzRkN2MzYTp7Im5hb
WUiOjF9', '2019-05-09 06:08:12.306625'),
('fq0czwxzas1e5bjz5cx5pr6ytm8uhejy',
'eyJ1c2VyaWQiOjExfQ:1mfKoD:eQbeRUgZ8NFqCleEdS6fE0NAoRs3zn6_B82
CZb4YtiQ', '2021-11-09 11:40:01.472965'),
('gq3vdjxoy34hxkorw2d8nztdusdknegl',
'eyJ1c2VyaWQiOjF9:1mpUvs:Lry5yh51WzsY8judWu-
ApNb05fEC5oHytMQ9bZSqbGU', '2021-12-07 12:29:56.098464'),

3
36

('gsqk7v1ei7yhuvcbxw6r8vxhgbuzz7zx',
'eyJ1c2VyaWQiOjJ9:1mkkhL:_T8wyqi-MJi-K7_a-0EPz-
h6HRouyRfpeAhHqTS5N04', '2021-11-24 10:19:19.270816'),
('h2up0dvopjvwswxnvprj7id9lgrivhus',
'eyJ1c2VyaWQiOjIzfQ:1mfM3f:FAuAUdY-
ly6qun6t571yt1pYKVGhXfjbjhiruld5rNs', '2021-11-09
13:00:03.703809'),
('hbv74sg6w6e4wp89vq807vw0xhkh5s1h',
'MzU0ZWYzNTQ3MjM4MWZlOTVjM2M1MWQ4MmE5ODE0OTlkNDRkNDkwMDp7InVzZ
XJpZCI6MX0=', '2020-01-10 07:40:38.067382'),
('hhtt48je70l9nzw6dee4ocuxxm9blqej',
'NGRhY2JkNmQ4ZTM4OTU0Y2UzMzFlZmZmOTgzYmE0MWVkOThiNjc2NTp7Im5hb
WUiOjEsInVzZXJpZCI6MX0=', '2019-05-09 10:12:38.380843'),
('hsb5814on7cph0wvy0yls67ca94ngcq3',
'eyJ1c2VyaWQiOjE5fQ:1mBzgz:cug3sAkQKH-bQBkB9O5l0UsDJL-
37eDV8mR9Qau3elA', '2021-08-20 13:15:17.195464'),
('i530ldontosd9c37rlmr7i190cc8j54c',
'eyJ1c2VyaWQiOjExfQ:1mfGFZ:PSpZPmdPYnGzwCScqY4tYBkDj8BMVATwweZ
jjxmG5dk', '2021-11-09 06:47:57.629900'),
('i77fui9jgj9yk7ncx7u4ph5d6kg0nl6c',
'eyJ1c2VyaWQiOjE5fQ:1mDPJC:kqt800XGsVGRjHS3TmeLFrJbrpIK4-
GbH4ZirwIc7S4', '2021-08-24 10:48:34.860638'),
('ic3hqykgws5iy6fz5ns6h6f921jbjzmt',
'eyJ1c2VyaWQiOjExfQ:1kywHL:I_tahJ0VJb7myAbMbXpWZu9XrSaAMmduNxG
d2x5gtmY', '2021-01-25 12:26:35.043761'),
('iz6wcyx97x1w6mpfc51g1tj72z2xghfn',
'eyJ1c2VyaWQiOjl9:1kwlIp:YKOKMwJARe6w057AKTGY1-
GCuRcZAeAbJ0bdQao23wY', '2021-01-19 12:19:07.663490'),
('jgcbya9z2s6b6mmldfv28lm18imc73m8',
'eyJ1c2VyaWQiOjIxfQ:1mDnYP:GTRQ2I-UYLdsCCyA0-
WsFSAVBNno1wLo6lk4M8JS0OU', '2021-08-25 12:41:53.312085'),
('jpkxxiej4bdjin5tpdjm0xqhdooexz9o',
'eyJ1c2VyaWQiOjExfQ:1mBEdk:YOk6fHHfBMmtt5ZvSyzgy13Az8JS59iXbU4
LO1Ps1RI', '2021-08-18 11:00:48.423036'),
('k7dyn4irgrj5wb4jucb4po527iw724dp',
'eyJ1c2VyaWQiOjEzfQ:1l0JrY:2_TJ4L_XoHdOW51Zdp0MOdyBEZEzntk5pdX
ZFDmX9x4', '2021-01-29 07:49:40.202848'),
('ktjsa2dwmkzggkc8htfro0m1zf2kt78d',
'eyJ1c2VyaWQiOjI0fQ:1mfiev:rUgpc2VOr-
8MQnmWwZsSVM_IqXQAA3Bacheqmp_LQ2o', '2021-11-10
13:08:01.503971'),
('kxla8qbe1hs1tb4a04ewpmdmi5fx1evc',
'eyJ1c2VyaWQiOjJ9:1mpVxM:dUv2r33P7CFNLdnAjy8jCk-
OJOlisN07sHd5Rf7zCLg', '2021-12-07 13:35:32.342260'),
('o7x1vhluuypdfmgv7fmv6nohgfn5ub55',
'NzMyZjlhNzFhZjk2ZGUzZmFiMmIzYjMwNTJkYTg5MDUzNmNlMDk4Mjp7InVzZ
XJpZCI6MTZ9', '2020-01-02 12:51:55.659179'),
('oc4pzt7ijx1rzj09m2ve6b6y5uwc6wt5',
'eyJ1c2VyaWQiOjF9:1mkk7W:6GDuPOPoLHMwOYLys3lXIgfM79Fq36HHW31-
utzbeyk', '2021-11-24 09:42:18.649798'),
('owqt9fqa6pkheboduh6f4k5p4lkwj0yc',
'eyJ1c2VyaWQiOjExfQ:1mfiXk:Wzn12pygxu_2Z1TzCSC4bKDawuXj_i7_BFL
hJjKx-10', '2021-11-10 13:00:36.684741'),
('psdjoq42u7lfqwfodftic5x6z9ij34nk',
'eyJ1c2VyaWQiOjExfQ:1mAXDq:a8YYY1YJU3jPv03qo9-
VcrjRHnDWRSqGseiR93n0GVM', '2021-08-16 12:39:10.518259'),

3
37

('q6hp9a2l9dbrclvox0o02x1aamx1ukj7',
'eyJ1c2VyaWQiOjIyfQ:1mfGz3:wkq7ZgyB738cK3Jugrc0viqb3eI1C0gWhyy
pHF_DE-A', '2021-11-09 07:34:57.860407'),
('qnaolidvfx6bu9ra3uyqvkgva7bv92f1',
'OTk3NTk2YTE0NjM5MWQ0OGQ0MjY3NzBjNzdhOTc0ZWJhM2ZkMzdkMjp7InVzZ
XJpZCI6MX0=', '2019-05-14 05:34:50.069335'),
('r0lswf36qh5gt0i2afmqqeu550j9vohq',
'eyJ1c2VyaWQiOjF9:1oOvxv:-
o0Ngd4h_vqB4DUhIuathTt7_CkrfaHFDCAmdaRTbwU', '2022-09-02
06:58:47.164716'),
('r9qk0kd407g591hugz99fhps8zofh69s',
'eyJ1c2VyaWQiOjE4fQ:1mBxnG:qkd9MTM_FhhghUpV90qngEkwkoSKYdLbfwR
KBLhK7Qg', '2021-08-20 11:13:38.356475'),
('rfq3uvadj7qsqrz7qlcyie9wscsqz1nr',
'eyJ1c2VyaWQiOjI1fQ:1mk2vF:mTIne3EU3rECWccrUfyPmy7XxIwdAzxhig4
S5oOngho', '2021-11-22 11:34:45.897995'),
('rn48bwukkb2yv60kvkacr8nc0njr5xky',
'eyJ1c2VyaWQiOjExfQ:1mfggv:xsbmrrGzxtrEFgspA0Wp7oWTp9qDl0shDSl
GG8fHJo4', '2021-11-10 11:01:57.519295'),
('s7ui2zx2cslubpch6dm7iaxlz2wlsdgg',
'eyJ1c2VyaWQiOjExfQ:1mk1Xz:fpol-krFazPxkK0b4gEoocpOXHcd-
eaYoxg26CU420Q', '2021-11-22 10:06:39.923928'),
('sdcvtwp7s5yj8q1lb0mdvlg8nj5wujqo',
'eyJ1c2VyaWQiOjEyfQ:1kzJ3p:0g6nRuJv3TXWVpANqNgbJcrUv96ZU5UQwv3
bgqBbL1I', '2021-01-26 12:46:09.538596'),
('tejgl09oettnyva23kqdbns5nfz5g8ug',
'OTk3NTk2YTE0NjM5MWQ0OGQ0MjY3NzBjNzdhOTc0ZWJhM2ZkMzdkMjp7InVzZ
XJpZCI6MX0=', '2019-05-09 11:19:24.387679'),
('tx26u0gjaebi1m6miqvms83dw452rw7c',
'eyJ1c2VyaWQiOjExfQ:1mg01H:i0OHhHK7t2WdfySWwMXXs92TaT7Vn1UQf2i
0eZBO70s', '2021-11-11 07:40:15.053714'),
('u5icgvq3qt5nthdlv99go3r804ccghbo',
'MmE4N2EzZmM3NTI1ODc3MjUxYjUxNWM3OWM4ZGExNWViMzRkN2MzYTp7Im5hb
WUiOjF9', '2019-05-09 06:00:13.573226'),
('ws2o4cq1jbqepe0e9s9v7n4erxatq9ic',
'eyJ1c2VyaWQiOjE1fQ:1l2CgI:SmlpAnZzplZhPTFJ_rkEJstnZRl2CYWyTca
h7PHPv-M', '2021-02-03 12:33:50.352453'),
('x5tpr0r7bu57jws6fpdwhi841252o6o2',
'eyJ1c2VyaWQiOjJ9:1oFxxT:uvC1LIN0lwgSyiMgEmEl1yH9W_bC1GTe5QDFs
BdOQqM', '2022-08-08 13:17:15.650911'),
('xc6si4gdotxq06mslnwwjtewdhzuyh44',
'eyJ1c2VyaWQiOjR9:1mnfuL:VMKdQ2gcb4wQppelEf-
wAdfTSSkA6nWPCeQODqD3NOM', '2021-12-02 11:48:49.418566'),
('zega5sz46ivu1tb1o1mtmg3v2ysxog1w',
'eyJ1c2VyaWQiOjh9:1kuVm4:L7RizVvw4EC0IyYCYAIhGjC8lvZol_Z1abqVw
dkdKkY', '2021-01-13 07:20:00.767751'),
('zq2idu1e3gkyelhzpa24k5jixqhj74da',
'eyJ1c2VyaWQiOjJ9:1mne5B:sOSzXh5WGI5se1746XNpUgE1_UZZw07Xzb03K
IcZXfY', '2021-12-02 09:51:53.585889'),
('zqfzqdolomzyjb8lckwouaztv90i58gi',
'eyJ1c2VyaWQiOjJ9:1mt6QE:e6TQ4CyC7rlTpZA0HyGK2zrDnv_EbGQrGGZUz
gCcnCM', '2021-12-17 11:08:10.160908');

--

3
38

-- Table structure for table

`remote_user_clientregister_model`
--

CREATE TABLE IF NOT EXISTS `remote_user_clientregister_model`

(
`id` int(11) NOT NULL AUTO_INCREMENT,
`username` varchar(30) NOT NULL,
`email` varchar(30) NOT NULL,
`password` varchar(50) NOT NULL,
`phoneno` varchar(50) NOT NULL,
`country` varchar(30) NOT NULL,
`state` varchar(30) NOT NULL,
`city` varchar(30) NOT NULL,
`address` varchar(3000) NOT NULL,
`gender` varchar(300) NOT NULL,
PRIMARY KEY (`id`)
) ENGINE=InnoDB DEFAULT CHARSET=latin1 AUTO_INCREMENT=3 ;

--
-- Dumping data for table `remote_user_clientregister_model`
--

INSERT INTO `remote_user_clientregister_model` (`id`,

`username`, `email`, `password`, `phoneno`, `country`,
`state`, `city`, `address`, `gender`) VALUES
(1, 'Arun', '[email protected]', 'Arun', '9535866270',
'India', 'Karnataka', 'Bangalore', '#782,4th
Cross,Rajajinagar', 'Male'),
(2, 'Manjunath', '[email protected]', 'Manjunath',
'9535866270', 'India', 'Karnataka', 'Bangalore', '#892,4th
Cross,Rajajinagar', 'Male');

--
-- Table structure for table
`remote_user_cyberbullying_detection_type`
--

CREATE TABLE IF NOT EXISTS

`remote_user_cyberbullying_detection_type` (
`id` int(11) NOT NULL AUTO_INCREMENT,
`Tweet_Message` varchar(30000) NOT NULL,
`Prediction` varchar(300) NOT NULL,
PRIMARY KEY (`id`)
) ENGINE=InnoDB DEFAULT CHARSET=latin1 AUTO_INCREMENT=9 ;

--
-- Dumping data for table
`remote_user_cyberbullying_detection_type`
--

INSERT INTO `remote_user_cyberbullying_detection_type` (`id`,

`Tweet_Message`, `Prediction`) VALUES

3
39

(1, 'Karma. I hope it bites Kat on the butt. She is just

nasty. #mkr', 'gender'),
(2, 'Karma. I hope it bites Kat on the butt. She is just
nasty. #mkr', 'gender'),
(3, 'even alle bully beatdown afleveringen gedownload van
season 1 en 2 :)', 'not_cyberbullying'),
(4, 'I’m sick of people still making jokes about rape
and using ‘gay’ as an insult. Just stop it.',
'gender'),
(5, 'the right thing to do? I as a Christian woman and a
Southern Baptist, strongly demand that you tell the truth.
Take responsibility. You have already lost millions of dollars
', 'religion'),
(6, 'RT @RiotFest: This is fucking terrible. People have no
respect anymore. https://t.co/GHPyGPnidl',
'other_cyberbullying'),
(7, 'know them niggas lookin so you know they gon hate
😈', 'ethnicity'),
(8, '@UmarFarooq_ radial momentum means that it does not fly
away. In the same way, both the plane and the atmosphere have
radial momentum.', 'not_cyberbullying');

--
-- Table structure for table `remote_user_detection_accuracy`
--

CREATE TABLE IF NOT EXISTS `remote_user_detection_accuracy` (

`id` int(11) NOT NULL AUTO_INCREMENT,
`names` varchar(300) NOT NULL,
`ratio` varchar(300) NOT NULL,
PRIMARY KEY (`id`)
) ENGINE=InnoDB DEFAULT CHARSET=latin1 AUTO_INCREMENT=69 ;

--
-- Dumping data for table `remote_user_detection_accuracy`
--

INSERT INTO `remote_user_detection_accuracy` (`id`, `names`,

`ratio`) VALUES
(64, 'MultinomialNB', '75.82555823461578'),
(65, 'SVM', '81.24541356536325'),
(66, 'Logistic Regression', '82.66065625327602'),
(67, 'Decision Tree Classifier', '78.2471957228221'),
(68, 'SGD Classifier', '83.65656777439983');

--
-- Table structure for table `remote_user_detection_ratio`
--

CREATE TABLE IF NOT EXISTS `remote_user_detection_ratio` (

`id` int(11) NOT NULL AUTO_INCREMENT,
`names` varchar(300) NOT NULL,

3
40

`ratio` varchar(300) NOT NULL,

PRIMARY KEY (`id`)
) ENGINE=InnoDB DEFAULT CHARSET=latin1 AUTO_INCREMENT=27 ;

--
-- Dumping data for table `remote_user_detection_ratio`
--

INSERT INTO `remote_user_detection_ratio` (`id`, `names`,

`ratio`) VALUES
(22, 'not_cyberbullying', '25.0'),
(23, 'gender', '37.5'),
(24, 'religion', '12.5'),
(25, 'other_cyberbullying', '12.5'),
(26, 'ethnicity', '12.5');

/*!40101 SET CHARACTER_SET_CLIENT=@OLD_CHARACTER_SET_CLIENT

*/;
/*!40101 SET CHARACTER_SET_RESULTS=@OLD_CHARACTER_SET_RESULTS
*/;
/*!40101 SET COLLATION_CONNECTION=@OLD_COLLATION_CONNECTION
*/;

4
41

5.3.2 OUTPUT SCREENS

Register page

Register form

4
42

User Login

Preduction page

4
43

Line chart

Pie chart

4
44

Ratio

4
45

Bar graph

Information Stored

4
46

5.3.3 RESULT ANALYSIS

A. MACHINE LEARNING AND DEEP LEARNING APPROACHES

In this section, the evaluation of DEA-RNN is performed over datasets crawled from
Twitter utilizing these metrics: recall, precision, F-measure, accuracy, and specificity.
The input dataset and the data annotation are described in sections III-A and III-B.
Two baseline cyberbullying models based on deep learning, namely Bi-LSTM [21],
RNN [21], and three baseline cyberbullying models based on machine learning
models, namely, SVM [26], Multinomial Naive Bayes (MNB) [11], and R [11] are
used for the comparison with the proposed DEA-RNN model. These models have
been selected from state-of-the-art cyberbullying detection in social media.

1. A. EVALUATION METRICS

This sub-section briefly highlights the evaluation metrics utilized in this study to
evaluate the efficiency of DEA-RNN. The evaluation process is performed based on
the following metrics: accuracy, recall, precision, F-measure, specificity and
computing training time. However, each method is run (N = 20) times for all
experiments to obtain an average of obtained results for each evaluation metric. These
performance metrics are described in Table 2.

4
47

2. EXPERIMENTAL RESULTS

This sub-section discusses the obtained experimental results of DEA-RNN classifier

in comparison with some considered baseline deep learning models, namely Bi-
LSTM, RNN, and other baseline machine learning models, namely MNB, RF, and
SVM. The prediction results of cyberbullying are validated based on various input
dataset scenarios 60:40% (Scenario 1), 70:30% (Scenario 2), and 90:10% (Scenario
3). The performance evaluation is carried out in terms of the aforesaid metrics. The
experiments were executed M = 20 times for each classifier over each dataset input
scenario. Then, the average of the performance metrics is computed using equations
as described in Table 2. The overall performance comparison results on various
classifiers over different dataset input scenarios are illustrated in Table 3

4
48

A. AVERAGE ACCURACY

The proposed DEA-RNN model is evaluated in terms of accuracy compared to

the considered existing models by computing the average accuracy for all
scenarios. As shown in Figure 3, the DEA-RNN model has obtained the highest
average accuracy of 90.45% in scenarios 3, while other methods such as Bi-
LSTM, RNN, SVM, MNB, and RF have got 88.74%, 87.15%, 85.21%, 82.26%,
and 83.45%, respectively. It is observed that the performance of deep learning
models (Bi-LSTM and RNN) is better than machine learning models (SVM, RF,
and MNB)

4
49

B.PERFORMANCE EVALUATION IN TERMS OF TRAINING TIME

The Training time of the proposed model was compared with baseline models.
Where, scenario 2 has been taken into consideration for computing the training time.
It can be observed that the proposed DEA-RNN model has less training time
compared to other deep learning Bi-LSTM, RNN baseline models. The training time
of Bi-LSTM is more than the proposed model, RNN as well as the machine learning
models, but the achievement of Bi-LSTM is better than the other baseline models and
less the proposed model.

4
50

6.TESTING

6.1 INTRODUCTION

The purpose of testing is to discover errors. Testing is the process of trying to

discover every conceivable fault or weakness in a work product. It provides a way to
check the functionality of components, sub assemblies, assemblies and/or a finished
product It is the process of exercising software with the intent of ensuring that the

Software system meets its requirements and user expectations and does not fail in an
unacceptable manner. There are various types of test. Each test type addresses a
specific testing requirement.

TYPES OF TESTS
The following are the Testing Methodologies:

o Unit Testing.
o Integration Testing.
o User Acceptance Testing.
o Output Testing.
o Validation Testing.

Unit Testing

Unit testing focuses verification effort on the smallest unit of Software design that is
the module. Unit testing exercises specific paths in a module’s control structure to

ensure complete coverage and maximum error detection. This test focuses on each
module individually, ensuring that it functions properly as a unit. Hence, the naming
is Unit Testing.

During this testing, each module is tested individually and the module
interfaces are verified for the consistency with design specification. All important
processing path are tested for the expected results. All error handling paths are also
tested.

Integration Testing

5
51

Integration testing addresses the issues associated with the dual problems of
verification and program construction. After the software has been integrated a set of
high order tests are conducted. The main objective in this testing process is to take
unit tested modules and builds a program structure that has been dictated by design.

The following are the types of Integration Testing:

1. Top Down Integration

This method is an incremental approach to the construction of program structure.
Modules are integrated by moving downward through the control hierarchy,
beginning with the main program module. The module subordinates to the main
program module are incorporated into the structure in either a depth first or breadth
first manner.
In this method, the software is tested from main module and individual stubs
are replaced when the test proceeds downwards.

3. Bottom-up Integration

This method begins the construction and testing with the modules at the lowest
level in the program structure. Since the modules are integrated from the bottom
up, processing required for modules subordinate to a given level is always
available and the need for stubs is eliminated. The bottom up integration strategy
may be implemented with the following steps:

 The low-level modules are combined into clusters into clusters that
perform a specific Software sub-function.
 A driver (i.e.) the control program for testing is written to coordinate
test case input and output.
 The cluster is tested.
 Drivers are removed and clusters are combined moving upward in the
program structure
The bottom up approaches tests each module individually and then each module is
module is integrated with a main module and tested for functionality.

User Acceptance Testing

5
52

User Acceptance of a system is the key factor for the success of any system. The
system under consideration is tested for user acceptance by constantly keeping in
touch with the prospective system users at the time of developing and making changes
wherever required. The system developed provides a friendly user interface that can
easily be understood even by a person who is new to the system.

Output Testing

After performing the validation testing, the next step is output testing of the proposed
system, since no system could be useful if it does not produce the required output in
the specified format. Asking the users about the format required by them tests the
outputs generated or displayed by the system under consideration. Hence the output
format is considered in 2 ways – one is on screen and another in printed format.

Validation Checking

Validation checks are performed on the following fields.

Text Field:

The text field can contain only the number of characters lesser than or equal to
its size. The text fields are alphanumeric in some tables and alphabetic in other
tables. Incorrect entry always flashes and error message.

Numeric Field:

The numeric field can contain only numbers from 0 to 9. An entry of any character
flashes an error messages. The individual modules are checked for accuracy and what
it has to perform. Each module is subjected to test run along with sample data. The
individually tested modules are integrated into a single system. Testing involves
executing the real data information is used in the program the existence of any
program defect is inferred from the output. The testing should be planned so that all
the requirements are individually tested.
A successful test is one that gives out the defects for the inappropriate data and
produces and output revealing the errors in the system.

Preparation of Test Data

5
53

Taking various kinds of test data does the above testing. Preparation of test data
plays a vital role in the system testing. After preparing the test data the system under
study is tested using that test data. While testing the system by using test data errors
are again uncovered and corrected by using above testing steps and corrections are
also noted for future use.

Using Live Test Data:

Live test data are those that are actually extracted from organization files. After
a system is partially constructed, programmers or analysts often ask users to key in a
set of data from their normal activities. Then, the systems person uses this data as a
way to partially test the system. In other instances, programmers or analysts extract a
set of live data from the files and have them entered themselves.

It is difficult to obtain live data in sufficient amounts to conduct extensive

testing. And, although it is realistic data that will show how the system will perform
for the typical processing requirement, assuming that the live data entered are in fact
typical, such data generally will not test all combinations or formats that can enter the
system. This bias toward typical values then does not provide a true systems test and
in fact ignores the cases most likely to cause system failure.
Using Artificial Test Data:

Artificial test data are created solely for test purposes, since they can be
generated to test all combinations of formats and values. In other words, the artificial
data, which can quickly be prepared by a data generating utility program in the
information systems department, make possible the testing of all login and control
paths through the program.

The most effective test programs use artificial test data generated by persons
other than those who wrote the programs. Often, an independent team of testers
formulates a testing plan, using the systems specifications.

The package “Virtual Private Network” has satisfied all the requirements
specified as per software requirement specification and was accepted.

6.2 USER TRAINING

5
54

Whenever a new system is developed, user training is required to educate them about
the working of the system so that it can be put to efficient use by those for
whom the system has been primarily designed. For this purpose the normal
working of the project was demonstrated to the prospective users. Its working
is easily understandable and since the expected users are people who have
good knowledge of computers, the use of this system is very easy.

6.3 MAINTAINENCE

This covers a wide range of activities including correcting code and design
errors. To reduce the need for maintenance in the long run, we have more accurately
defined the user’s requirements during the process of system development. Depending
on the requirements, this system has been developed to satisfy the needs to the largest
possible extent. With development in technology, it may be possible to add many
more features based on the requirements in future. The coding and designing is simple
and easy to understand which will make maintenance easier.

TESTING STRATEGY:

A strategy for system testing integrates system test cases and design techniques
into a well planned series of steps that results in the successful construction of
software. The testing strategy must co-operate test planning, test case design, test
execution, and the resultant data collection and evaluation .A strategy for software
testing must accommodate low-level tests that are necessary to verify that a small
source code segment has been correctly implemented as well as high level tests
that validate major system functions against user requirements.

Software testing is a critical element of software quality assurance and represents

the ultimate review of specification design and coding. Testing represents an
interesting anomaly for the software. Thus, a series of testing are performed for the
proposed system before the system is ready for user acceptance testing.

5
55

6.2 SYSTEM TESTING:

Software once validated must be combined with other system elements (e.g.
Hardware, people, database). System testing verifies that all the elements are proper
and that overall system function performance is

achieved. It also tests to find discrepancies between the system and its original
objective, current specifications and system documentation.

UNIT TESTING:

In unit testing different are modules are tested against the specifications
produced during the design for the modules. Unit testing is essential for verification of
the code produced during the coding phase, and hence the goals to test the internal
logic of the modules. Using the detailed design description as a guide, important
Conrail paths are tested to uncover errors within the boundary of the modules. This
testing is carried out during the programming stage itself. In this type of testing step,
each module was found to be working satisfactorily as regards to the expected output
from the module.

Integration testing

Integration tests are designed to test integrated software components to

determine if they actually run as one program. Testing is event driven and is more
concerned with the basic outcome of screens or fields. Integration tests demonstrate
that although the components were individually satisfaction, as shown by successfully
unit testing, the combination of components is correct and consistent. Integration
testing is specifically aimed at exposing the problems that arise from the
combination of components.

5
56

Functional test
Functional tests provide systematic demonstrations that functions tested are
available as specified by the business and technical requirements, system
documentation, and user manuals.

Functional testing is centered on the following items:

Valid Input : identified classes of valid input must be accepted.

Invalid Input : identified classes of invalid input must be rejected.

Functions : identified functions must be exercised.

Output : identified classes of application outputs must be

exercised. Systems/Procedures: interfacing systems or procedures must be

invoked.

Organization and preparation of functional tests is focused on requirements, key

functions, or special test cases. In addition, systematic coverage pertaining to identify
Business process flows; data fields, predefined processes, and successive processes
must be considered for testing. Before functional testing is complete, additional tests
are identified and the effective value of current tests is determined.

System Test
System testing ensures that the entire integrated software system meets
requirements. It tests a configuration to ensure known and predictable results. An
example of system testing is the configuration oriented system integration test.
System testing is based on process descriptions and flows, emphasizing pre-driven
process links and integration points.

White Box Testing

White Box Testing is a testing in which in which the software tester has

5
knowledge of the inner workings, structure and language of the software, or at least its

5
57

purpose. It is purpose. It is used to test areas that cannot be reached from a black box
level.

Black Box Testing

Black Box Testing is testing the software without any knowledge of the inner
workings, structure or language of the module being tested. Black box tests, as most
other kinds of tests, must be written from a definitive source document, such as
specification or requirements document, such as specification or requirements
document. It is a testing in which the software under test is treated, as a black box
.you cannot “see” into it. The test provides inputs and responds to outputs without
considering how the software works.

5
58

7.VALIDATIOn

7.1 INTRODUCTION

Validation are performed on the following fields.

Text Field:

The text field can contain only the number of characters lesser than or equal to
its size. The text fields are alphanumeric in some tables and alphabetic in other
tables. Incorrect entry always flashes and error message.
Numeric Field:

The numeric field can contain only numbers from 0 to 9. An entry of any
character flashes an error messages. The individual modules are checked for accuracy
and what it has to perform. Each module is subjected to test run along with sample
data. The individually tested modules are integrated into a single system. Testing
involves executing the real data information is used in the program the existence of
any program defect is inferred from the output. The testing should be planned so
that all the requirements are individually tested.
A successful test is one that gives out the defects for the inappropriate data
and produces and output revealing the errors in the system.

Preparation of Test Data

Taking various kinds of test data does the above testing. Preparation of test
data plays a vital role in the system testing. After preparing the test data the system
under study is tested using that test data. While testing the system by using test data
errors are again uncovered and corrected by using above testing steps and corrections
are also noted for future use.

Using Live Test Data:

Live test data are those that are actually extracted from organization files.

After a system is partially constructed, programmers or analysts often ask users to key

in a set of data from their normal activities. Then, the systems person uses this data as

5
59

a way to partially test the system. In other instances, programmers or analysts extract

a set of live data from the files and have them entered themselves.

It is difficult to obtain live data in sufficient amounts to conduct extensive

testing. And, although it is realistic data that will show how the system will perform
for the typical processing requirement, assuming that the live data entered are in fact
typical, such data generally will not test all combinations or formats that can enter the
system. This bias toward typical values then does not provide a true systems test and
in fact ignores the cases most likely to cause system failure.
Using Artificial Test Data:

Artificial test data are created solely for test purposes, since they can be generated to
test all combinations of formats and values. In other words, the artificial data, which
can quickly be prepared by a data generating utility program in the information
systems department, make possible the testing of all login and control paths through
the program.

The most effective test programs use artificial test data generated by persons other
than those who wrote the programs. Often, an independent team of testers formulates
a testing plan, using the systems specifications.

The package “Virtual Private Network” has satisfied all the requirements specified as
per software requirement specification and was accepted.

USER TRAINING

Whenever a new system is developed, user training is required to educate them about
the working of the system so that it can be put to efficient use by those for whom the
system has been primarily designed. For this purpose the normal working of the
project was demonstrated to the prospective users. Its working is easily
understandable and since the expected users are people who have good knowledge of
computers, the use of this system is very easy.

MAINTAINENCE

This covers a wide range of activities including correcting code and design errors. To
reduce the need for maintenance in the long run, we have more accurately defined the

6
60

user’s requirements during the process of system development. Depending on the

requirements, this system has been developed to satisfy the needs to the largest
possible extent. With development in technology, it may be possible to add many
more features based on the requirements in future. The coding and designing is simple
and easy to understand which will make maintenance easier.

TESTING STRATEGY :

A strategy for system testing integrates system test cases and design techniques into a
well planned series of steps that results in the successful construction of software.
The testing strategy must co-operate test planning, test case design, test execution,
and the resultant data collection and evaluation .A strategy for software testing must
accommodate low-level tests that are necessary to verify that a small source code
segment has been correctly implemented as well as high level tests that validate
major system functions against user requirements.

Software testing is a critical element of software quality assurance and represents the
ultimate review of specification design and coding. Testing represents an interesting
anomaly for the software. Thus, a series of testing are performed for the proposed
system before the system is ready for user acceptance testing.

SYSTEM TESTING:

Software once validated must be combined with other system elements (e.g.
Hardware, people, database). System testing verifies that all the elements are proper
and that overall system function performance is achieved. It also tests to find
discrepancies between the system and its original objective, current specifications
and system documentation.

UNIT TESTING:

In unit testing different are modules are tested against the specifications produced
during the design for the modules. Unit testing is essential for verification of the code
produced during the coding phase, and hence the goals to test the internal logic of
the modules. Using the detailed design description as a guide, important Conrail
paths are tested to uncover errors within the boundary of the modules. This testing is
carried out during the programming stage itself. In this type of testing step, each

6
61

module was found to be working satisfactorily as regards to the expected output from
the module.

In Due Course, latest technology advancements will be taken into

consideration. As part of technical build-up many components of the networking
system will be generic in nature so that future projects can either use or interact with
this. The future holds a lot to offer to the development and refinement of this project.

Different categories for current phishing attacks were identified by using a

domain frequency analysis. While standalone and compromised domains were the
most common approaches, free hosting services, cloud web servers and malware blog
posts represent many current phishing attacks due to their cost and effectiveness for
phishing campaigns.

7.2 CONCLUSION

Finally, we demonstrated that machine learning models using handcrafted

URL features decreased their performance over time, up to 10:42% accuracy in the
case of the Light GBM algorithm from the year 2016 to 2020. For this reason,
machine learning methods should be trained with recent URLs to prevent substantial
ageing from the date of its release. In the future, we will add more information about
the samples into the analysis, such as the source code of the website and a screenshot
of its content, which could be useful to increase the phishing detection performance.
In addition, we will enlarge our dataset, including such information. Finally,
observing that deep learning techniques and automatic feature extraction obtained
promising results over traditional feature extraction, we intend to explore different
URL codifications to improve detection performance

6
62

8.CONCLUSION

This paper developed an efficient tweet classification model to enhance the

effectiveness of topic models for the detection of cyber-bullying events. DEA RNN
was developed by combining both the DEA optimization and the Elman type RNN for
efficient parameter tuning. Furthermore, it was tested in comparison with the existing
Bi-LSTM, RNN, SVM, RF, and MNB methods on a newly created Twitter dataset,
which was extracted using CB keywords. The experimental analysis showed that the
DEA-RNN had achieved optimal results compared to the other existing methods in all
the scenarios with various metrics such as accuracy, recall, F-measure, precision, and
specificity. This signifies the impact of DEA on the performance of RNN. Although
the hybrid proposed model obtained higher performance rates than the other
considered existing models, the feature compatibility of DEA-RNN reduces when the
input data is increased greater than the initial input. The current study was limited
only to the Twitter dataset exclusively; other Social Media Platforms (SMP) such as
Instagram, Flickr, YouTube, Face book, etc., should be investigated in order to detect
the trend of cyber bullying. Then, the possibility of utilizing multiple source data for
cyber-bullying detection will be investigated in the future. Furthermore, we performed
the analysis only on the content of tweets; we could not perform the analysis in
relation to the users' behavior. This will be in future works. The proposed model
works to detect cyber bullying utilizing textual content of tweets, whereas theother
type of media such as images, video, and audio is still an open research area and
future research directions. Besides, we aim to classify and detect CB tweets in a real-
time stream.

6
63

REFERENCE

[1] F. Mishna, M. Khoury-Kassabri, T. Gadalla, and J. Daciuk, ``Risk

factors for involvement in cyber bullying: Victims, bullies and
bully_victims,'' Children Youth Services Rev., vol. 34, no. 1, pp. 63_70, Jan.
2012, doi: 10.1016/j.childyouth.2011.08.032.
[2] K. Miller, ``Cyberbullying and its consequences: How cyberbullying
is contorting the minds of victims and bullies alike, and the law's limited
available redress,'' Southern California Interdiscipl. Law J., vol. 26, no. 2,
p. 379, 2016.
[3] A. M. Vivolo-Kantor, B. N. Martell, K. M. Holland, and R. Westby,
``A systematic review and content analysis of bullying and cyber-bullying
measurement strategies,'' Aggression Violent Behav., vol. 19, no. 4,
pp. 423_434, Jul. 2014, doi: 10.1016/j.avb.2014.06.008.
[4] H. Sampasa-Kanyinga, P. Roumeliotis, and H. Xu, ``Associations
between cyberbullying and school bullying victimization and suicidal
ideation,
plans and attempts among Canadian schoolchildren,'' PLoS ONE, vol. 9,
no. 7, Jul. 2014, Art. no. e102145, doi: 10.1371/journal.pone.0102145.
[5] M. Dadvar, D. Trieschnigg, R. Ordelman, and F. de Jong,
``Improving cyberbullying detection with user context,'' in Proc. Eur.
Conf. Inf. Retr., in Lecture Notes in Computer Science: Including
Subseries Lecture Notes in Arti_cial Intelligence and Lecture Notes in
Bioinformatics, vol. 7814, 2013, pp. 693_696.
[6] A. S. Srinath, H. Johnson, G. G. Dagher, and M. Long,
``BullyNet: Unmasking cyberbullies on social networks,'' IEEE Trans.
Computat. Social Syst., vol. 8, no. 2, pp. 332_344, Apr. 2021, doi:
10.1109/TCSS.2021.3049232.
[7] A. Agarwal, A. S. Chivukula, M. H. Bhuyan, T. Jan, B. Narayan,
and M. Prasad, ``Identi_cation and classi_cation of cyberbullying
posts: A recurrent neural network approach using under-sampling and
class weighting,'' in Neural Information Processing (Communications
in Computer and Information Science), vol. 1333, H. Yang, K. Pasupa,
A. C.-S. Leung, J. T. Kwok, J. H. Chan, and I. King, Eds. Cham,

6
64

Switzerland: Springer, 2020, pp. 113_120.

[8] Z. L. Chia, M. Ptaszynski, F. Masui, G. Leliwa, and M. Wroczynski,
``Machine learning and feature engineering-based study into sarcasm
and irony classi_cation with application to cyberbullying detection,''
Inf. Process. Manage., vol. 58, no. 4, Jul. 2021, Art. no. 102600, doi:
10.1016/j.ipm.2021.102600.
[9] N. Yuvaraj, K. Srihari, G. Dhiman, K. Somasundaram, A. Sharma,
S. Rajeskannan, M. Soni, G. S. Gaba, M. A. AlZain, and M. Masud,
``Nature-inspired-based approach for automated cyberbullying classi_cation
on multimedia social networking,'' Math. Problems Eng., vol. 2021,
pp. 1_12, Feb. 2021, doi: 10.1155/2021/6644652.
[10] B. A. Talpur and D. O'Sullivan, ``Multi-class imbalance in text classi
_cation: A feature engineering approach to detect cyberbullying in
Twitter,'' Informatics, vol. 7, no. 4, p. 52, Nov. 2020, doi:
10.3390/informatics7040052.
[11] A. Muneer and S. M. Fati, ``A comparative analysis of machine
learning techniques for cyberbullying detection on Twitter,'' Futur. Internet,
vol. 12, no. 11, pp. 1_21, 2020, doi: 10.3390/_12110187.
[12] R. R. Dalvi, S. B. Chavan, and A. Halbe, ``Detecting a Twitter
cyberbullying using machine learning,'' Ann. Romanian Soc. Cell Biol., vol. 25,
no. 4,
pp. 16307_16315, 2021.