FRAUD PREVENTION IN REAL TIME

MOBILE SOCIAL NETWORKS

A Project Report
Submitted to Jawaharlal Nehru Technological University,
Kakinada in partial fulfillment for the award of the degree of

BACHELOR OF TECHNOLOGY
IN
COMPUTER SCIENCE & ENGINEERING
SUBMITTED BY

P. AMULYA 19MA1A0521
D. GAYATHRI 19MA1A0509
M. SIVANI 19MA1A0517
Y.V.L. SWETHA 19MA1A0531

Under the esteemed guidance of

Ms. A. VENNELA, M.Tech,

Assistant professor.

DEPARTMENTOF COMPUTER SCIENCE& ENGINEERING

QIS INSTITUTE OF TECHNOLOGY

(Approved by A.I.C.T.E, New Delhi & Affiliated to J.N.T.U, Kakinada)

(An ISO 9001:2015 Certified Institution)

VENGAMUKKAPALEM, ONGOLE- 523272, A.P.

2019-2023
 QIS INSTITUTE OF TECHNOLOGY
(Approved by AICTE, New Delhi & Affiliated to JNTU Kakinada)

(An ISO 9001:2015 Certified Institution)

VENGAMUKKAPALEM, ONGOLE-523272, A.P, INDIA

DEPARTMENT OF COMPUTER SCIENCE & ENGINEERING

BONAFIDE CERTIFICATE

This is to certify that the Project work entitled " FRAUD PREVENTION IN REAL
TIME MOBILE SOCIAL NETWORKS” is a bonafied work done by P.AMULYA
(19MA1A0521), D. GAYATHRI (19MA1A0509), M. SIVANI (19MA1A0517), Y. VENKATA
LAKSHMI SWETHA (19MA1A0531), in partial fulfillment of the requirement for the award of
the degree of Bachelor of Technology in COMPUTER SCIENCE ANDENGINEERING for the
academic year 2022-2023. This work is done under our supervision and guidance.

Signature of the Project Guide Signature of the Head of the Dept

Ms. A. VENNELA, M. Tech Mr. J. HARIBABU, M. Tech
ASSISTANT PROFESSOR ASSISTANT PROFESSOR

Signature of the External Examiner

Date:
 ACKNOWLEDGEMENT

“Task Successful” makes everyone happy. But the happiness will be gold without glitter if we
didn’t state the persons who supported us to make it a success.

We would like to place on record the deep sense of gratitude to the Honorable Secretary &
Correspondent Dr. N. SURYA KALYAN CHAKRAVATHY GARU, PhD, QIS GROUP Of
INSTITUTIONS, Ongole for providing necessaryfacilities to carrythe project work.

We express our gratitude to the Hon’ble Chairman Sri. N. NAGESWARA RAO GARU, QIS
GROUP OF INSTITUTIONS, Ongole for his valuable suggestions and advices in the B. Tech Course.

We express our gratitude to Dr. Y. SESHA RAO GARU, MTech, PhD, Principal of QIS Institute
Of Technology, Ongole for his valuable suggestions and advices in B. Tech course.

We express our gratitude to the HEAD OF THE DEPARTMENT of CSE, Mr. J.HARIBABU,
M.Tech, Assistant Professor, QIS Institute Of Technology, Ongole for his Constant supervision, guidance

and co-operation throughout the project.

We would like to express our thankfulness to our project guide Ms. A. VENNELA, M.Tech,
Assistant Professor, QIS Institute Of Technology, Ongole for her constant motivation and valuable help
throughout the project work.

Finally, we would like to thank our Parents, family and for their co-operation to complete this
project.

TEAM MEMBERS

P. AMULYA 19MA1A0521
D. GAYATHRI 19MA1A0509
M. SIVANI 19MA1A0517
Y. V.L SWETHA 19MA1A0531
 DECLARATION

We hereby declare that the work, being presented in this dissertation, entitled “FRAUD
PREVENTION IN REAL TIME MOBILE SOCIAL NETWORKS” submitted to
Jawaharlal Nehru Technological University Kakinada in partial fulfillment of requirements forthe award
of the degree of Bachelor of The Technology is carried out under the supervision of Ms. A. VENNELA,
M.Tech, Assistant Professor, Department of Computer Science & Engineering, QISIT, Ongole is
of my own and has not been submitted to any other University or Educational Institution for any
degree or diploma.

BY

P. AMULYA 19MA1A0521
D. GAYATHRI 19MA1A0509
M. SIVANI 19MA1A0517
Y.V.L SWETHA 19MA1A0531
 TABLE OF CONTENTS

CHAPTER TITLE PAGE NO.

ABSTRACT
1 INTRODUCTION 1-3
1.1 Fraud Information Diffusion Model
1.2 Dynamic Allocation of Strategy
1.3 Simulation Experiments on Datasets

2 LITERATURE SURVEY 4-5

3 SYSTEM STUDY 6
3.1 Feasibility Study
3.1.1 Economical Feasibility
3.1.2 Technical Feasibility
3.1.3 Social Feasibility
4 SYSTEM ANALYSIS 7-8
4.1 Existing System
4.1.1 Disadvantages
4.2 Proposed System
4.2.1 Advantages

5 SYSTEM REQUIREMENTS 9
5.1 Hardware Requirements
5.2 Software Requirements
6 SYSTEM DESIGN 10-17
6.1 System Architecture
6.1.1 Data flow Diagram
6.1.2 Data Owner
6.1.3 Data User
6.2 Modules Description
6.3 UML Diagrams
6.3.1 Use Case diagram
6.3.2 Class Diagram
6.3.3 Sequence Diagram
7 SYSTEM IMPLEMENTATION 18-31
7.1 Software Environment
7.1.2 General J2ME architecture
7.1.3 Building j2ME apps is the next stage
7.2 Source Code
8 SYSTEM TESTING 32-34
8.1 Testing
8.1.1 The method of assessment
8.1.2 Implementation and Evaluation results
8.1.3 A test of utility
8.2 Black Box Testing
9 RESULTS 35-38
9.1 Starting Home Page
9.2 Admin Login Page
9.3 Admin Page View
9.4 Add filter details and existing filter details
9.5 User Login Page
 9.6 User registration form
9.7 User page view
9.8 post uploading process

10. CONCLUSION 39

11. REFERENCES 40-42

 LIST OF FIGURES

Figure No Figure Name Page No

6.1 System Architecture 10

6.2 Data Flow diagram 11

9.1 Use case Diagram 15

9.2 Class Diagram 16

9.3 Sequence Diagram 17
 ABBREVATIONS

MSN - Mobile Social Networks

SWIR - Short Wave Infrared

SIR - Susceptible Infected Recovered

AOTA - Average Online Task Assignment

LOTA - Largest Online Task Assignment

OOD - Online Opinion Dynamics

UML - Unified Modelling Language

 ABSTRACT
Through the use of mobile devices, people can access real-time information services
provided by mobile social networks (MSNs). MSNs have significant levels of transparency
and autonomy, yet they have been plagued by widespread misinformation, fraud, and other
forms of abuse as a result. Limiting the availability of fraud alerts is crucial for preventing
further damage. How can scarce resources be best put to use while still allowing individuals
to suffer as little financial loss as possible as a result of fraudulent data? As a result, we
formulate the challenge of controlling fraudulent data as an optimal control problem, with
the overall cost being the sum of the control resources expended in enacting control
techniques and the losses incurred by people. We develop the best dynamic allocation of
control strategies based on the idea of optimum control. Additionally, we build a dynamics
model for fraud information diffusion by factoring in individuals' unpredictability of mind,
and we explore the pattern of fraud information diffusion and the stability of the dynamics
model.
 1. INTRODUCTION
With the boom of the Internet and the rapid popularization of intelligent mobile
devices, mobile social networks (MSNs) have grown up to become an important platform for
information dissemination [1]. MSNs can provide people with a variety of real-time information
services and have already penetrated into our daily life. The Internet-based MSNs have
exhibited their great charm and broad prospect in many application fields, such as instant
communication, life service, interactive entertainment, etc., and have attracted extensive
attention of the industry and the academia [2], [3]. However, the development of MSNs is like
a double-edged sword [4], [5]. When MSNs are increasingly becoming an indispensable part
of people’s lives, a series of unhealthy phenomena, such as fake news, rumors, online
promotion, and fraudulent activities are becoming more and more rampant, which pose a
serious threat on the normal social network activities [6], [7]. Besides, by means of the
emerging technologies of intelligent terminals, wireless networks, and online payment in
recent years, the high rate of fraud has caused great losses to people [8]. According to the
official data released by the security ministry, telecommunications fraud in MSNs has grown
at an annual rate of 20%–30% [9]. The following are two representative scenarios
Scenario A: One scenario is the Veracruz incident in August 2015 [10].A piece of
rumor saying “shootouts and kidnappings by drug gangs happening near schools in Veracruz”
spread in Twitter and Face book. This rumor caused severe chaos in the city and many serious
car crashes happened amid the hysteria.
Scenario B: Another shocking scenario occurred in August 2016 when a Chinese
university professor suffered a telecommunication-based fraud, lead into a serious loss of 17.6
million Yuan [11]. Criminals fabricated an elaborate hoax, used the network to transmit fraud
information and perform remote fraud victims.
Fraud information diffusion has become a prominent problem in social networks [12].
Those evidence highlight that effectively controlling the fraud information in MSNs
applications is of great significance. Here, we define so-called fraud information as a piece of
malicious information or false information, which aims to intentionally cause adverse effects,
such as mass panic or defraud victims of their property. In order to cope up with the spread of
such information in MSNs more effectively, it is an urgent need to study the pattern of fraud
informationdiffusion and further put forwardthe corresponding control measures.
Previously, some mathematical models have been used to model the diffusion
evolutionary process of fraud information in the network. Most of these models are based on
the theory of biological infectious disease because the spread process of infectious diseases in
biology and the diffusion process of fraud information in the network are very similar [13] .

Fraud Prevention In Real Time Mobile Social Networks Page No:1

 The most widely used model is the susceptible-infected recovered (SIR) model, in
which all individuals are divided into three categories: 1) susceptible; 2) infected; and 3)
recovered [14]. From the perspective of information diffusion, the semantics of susceptible,
infected, and recovered can fully correspond to the process of fraud information diffusion. If
an individual has not yet received any fraud information, it belongs to the susceptible state. If
an individual received fraud information and was misled, it belongs to the infected state. Ifan
individual was ever infected and now no longer believes the fraud information, it belongs tothe
recovered state.
Although the existing SIR-based derivation models can correctly describe the
transitional relationship and the dynamic evolutionary processes of node states, the spread of
fraud information in MSNs shows some new characteristics. First, the information sender and
receiver are human beings, and human mental activities are often complex. For example, the
individual will likely experience a series of mental activities, such as thinking, hesitating, and
wandering when receiving a piece of information [15], [16]. Second, the fraud information
diffusion processes in MSNs are the complex results of the continuous interactions of nodes
in different states [17]. Third, because of the psychological effect, repeated reception of the
same information may give users the feeling of disgust and lead to reverse psychology. The
data analysis about
4.4 million Twitter messages diffusion shows that in the process of information diffusion,
users will deviate from the original intention of information and produce the phenomenon of
emotional transfer [18].
Besides establishing dynamics models and revealing fraud information diffusion
laws, our ultimate goal should be to effectively control the diffusion of fraud information.
However, the implementation of any control or intervention for the system will incur a certain
“price” [19]. As for the process of controlling fraud information diffusion in MSNs, some
operational control measures will inevitably consume a certain amount of precious manpower
and material resources. For example, in response to the crisis of fraud information, the
government constantly sends authoritative messages to the network to prevent individuals
from being misled by it. All this need to cost a lot of limited communication and other
resources. Therefore, how to efficiently utilize limited control resources and minimize losses
of individuals by adopting proper control strategies have become an urgent issue to address.
Some of the existing research works can control the diffusion of fraud information to
some extent, but there are still some obvious issues [20],[21],[22]. The first issue that the they
usually adopts a single continuous or pulse control strategy,and mostly do not consider the

implementation efficiency of the control strategy and the utilization efficiency ofthe control.

Fraud Prevention In Real Time Mobile Social Networks Page No2

 The second issue is that while some works have realized the constraint of control
resources and transformed the control problem into the optimal dynamic allocation of control
resources, they ignore the harm of fraud information diffusion to individuals.
In order to overcome the above limitations, in this paper, we put forward a novel
dynamics model, called SWIR, which can accurately describe the dynamic process of fraud
information diffusion. Importantly, for the sake of efficiently utilizing the limited resources and
minimizing the losses of individuals, we establish the optimal control system to solve the
optimal dynamic allocation problem of control strategies for fraud information diffusion. The
main contributions of this paper are summarized as follows.
1.1 Fraud Information Diffusion Model: In consideration of the uncertain mental state of
individuals and the transitional relationship of individuals in different states, we establish the
SWIR model. It can more effectively describe the dynamic diffusion process of fraud
information in MSNs. In addition, we theoretically analyze the stability of the SWIR model and
the trend of fraud information diffusion.
1.2 Dynamic Allocation of the Control Strategies: In order to efficiently utilize
limited control resources and minimize losses of individuals caused by fraud information, we
propose to synergistic control strategies. We take the control resources consumption and the
losses of individuals as the total cost constraint. Then, we formulate the optimal control problem
to minimize the total cost, and model the control strategies as functions varying over time.
Finally, based on the optimal control theory, the optimal distribution of the control strategies
functions over time is derived.
1.3 Simulation Experiments on Datasets: We validate the correctness and efficiency of the
proposed diffusion model and the optimal control strategies on both synthetic datasets and real
social network datasets. The results demonstrate that our proposed diffusion model can
accurately describe the dynamic diffusion process of fraud information and our proposed
control strategies can effectively inhibit the fraud information in MSNs. In particular, the
optimal dynamic allocation control strategies can achieve minimum control resources
consumption and losses of individuals.
The rest of this paper is organized as follows. In Section II, some previous works
are reviewed. In Section III, we first establish a novel dynamics model of the fraud
informationdiffusion in MSNs. Then, we analyze the trend offraud information diffusion and
the stability of the dynamics model. Consequently, we propose two synergistic control
strategies to suppress the spread of fraud information and derive the optimal distribution of
the control strategies. The extensive simulations are conducted in Section IV. Section V
concludes this project.

Fraud Prevention In Real Time Mobile Social Networks Page No:3

 2. LITERATURE SURVEY

Online task assignment for crowdsensing in predictable mobile social

networks
Authors: M. Xiao, J. Wu, L. Huang, R. Cheng, and Y. Wang
Mobile crowdsensing is a new paradigm in which a crowd of mobile users exploit
their carried smart phones to conduct complex sensing tasks. In this paper, we focus on the
make span sensitive task assignment problems for the crowd sensing in mobile social
networks, where the mobility model is predicable, and the time of sending tasks and recycling
results is non-negligible. To solve the problems, we propose an Average make span sensitive
Online Task Assignment (AOTA) algorithm and a Largest make span sensitive Online Task
Assignment (LOTA) algorithm. In AOTA and LOTA, the online task assignments are
viewed as multiple rounds of virtual offline task assignments. Moreover, a greedy strategy
of small-task-first-assignment and earliest-idle- user-receive-task is adopted for each round
of virtual offline task assignment in AOTA, while the greedy strategy of large-task-first-
assignment and earliest- idle-user-receive-task is adopted for the virtual offline task
assignments in LOTA. Based on the two greedy strategies, both AOTA and LOTA can
achieve nearly optimal online decision performances. We prove this and give the competitive
ratios of the two algorithms. In addition, we also demonstrate the significant performance of
the two algorithms through extensive simulations, based on four real MSN traces and a
synthetic MSN trace.
Predicting the evolution of hot topics: A solution based on the online
opinion dynamics model in social network
Authors: L. Jiang, J. Liu, D. Zhou, Q. Zhou, X. Yang, and G. Yu
Predicting and utilizing the evolution trend of hot topics is critical for contingency
management and decision-making purposes of government bodies and enterprises. This
paper proposes a model named online opinion dynamics (OODs) where any node in a social
network has its unique confidence threshold and influence radius. The nodes in the OOD are
mainly affected by their neighbors and are also randomly influenced by unfamiliar nodes. In
the traditional opinion model, however, each node is affected by all other nodes, including its
friends. Furthermore, many traditional opinion evolution approaches are reviewed to see if all
nodes (participants) can eventually reach a consensus. On the contrary, OOD is more focused
on such details as concluding the overall trend of events and evaluating the support level of
each participant through numerical simulation.

Fraud Prevention In Real Time Mobile Social Networks Page No:4

 Experiments show that OOD is superior to the improvement of the original
Hegselmann-Krause (HK) model, HK-13 and HK-17, with respect to qualitative predictions
of the evolution trend of an event. The quantitative predictions of the HK model cannot be
used to make decisions, whereas the results of the OOD model are proved to be acceptable.
An on-demand coverage based self-deployment algorithm for big data
perception in mobile sensing networks
Authors: Y. Lin et al
Mobile Sensing Networks have been widely applied to many fields for big data
perception such as intelligent transportation, medical health and environment sensing.
However, in some complex environments and unreachable regions of inconvenience for
human, the establishment of the mobile sensing networks, the layout of the nodes and the
control of the network topology to achieve high performance sensing of big data are
increasingly becoming a main issue in the applications of the mobile sensing networks. To
deal with this problem, we propose a novel on-demand coverage based self- deployment
algorithm for big data perception based on mobile sensing networking this project. Firstly,
by considering characteristics of mobile sensing nodes, we extend the cellular automata
model and propose a new mobile cellular automata model for effectively characterizing the
spatial–temporal evolutionary process of nodes. Secondly, based on the learning automata
theory and the historical information of node movement, we further explore a new mobile
cellular learning automata model, in which nodes can self-adaptively and intelligently decide
the best direction of movement with low energy consumption. Finally, we propose a new
optimization algorithm which can quickly solve the node self-adaptive deployment problem,
thus, we derive the best deployment scheme of nodes in a short time. The extensive simulation
results show that the proposed algorithm in this paper outperforms the existing algorithms by
as much as 40% in terms of the degree of satisfaction ofnetwork coverage, the iterations of the
algorithm, the average moving steps of nodes and the energy consumption of nodes. Hence,
we believe that our work will make contributions to large-scale adaptive deployment and high
performance sensing scenarios of the mobile sensing networks. Quickly solve the node self-
adaptive deployment problem, we derive the best deployment scheme of nodes in a short time.
The extensive simulation results show that the proposed algorithm in this project out performs
the existing algorithms by as much as 40% in terms of the degree of satisfaction of network
coverage, the iterations of the algorithm, the average moving steps of nodes and the energy
consumption of nodes. Hence, we believe that our work will make contributions to large-
scale adaptive deployment and high performance sensing scenarios of the mobile sensing
networks.

Fraud Prevention In Real Time Mobile Social Networks Page No:5

 3. SYSTEM STUDY

3.1 Feasibility study:

The feasibility of the project is analyzed in this phase and business proposal is put forth
with a very general plan for the project and some cost estimates. During system analysis the
feasibility study of the proposed system is to be carried out. This is to ensure that the proposed
system is not a burden to the company. For feasibility analysis, some understanding ofthe major
requirements for the system is essential.

Three key considerations involved in the feasibility analysis are

 Economical feasibility
 Technical feasibility
 Social feasibility

3.1.1 Economical feasibility:

This study is carried out to check the economic impact that the system willhave on the
organization. The amount of fund that the company can pour into the research and development
of the system is limited. The expenditures must be justified. Thus the developed system as well
within the budget and this was achieved because most of the technologies used are freely
available. Only the customized products had to be purchased.
3.1.2 Technical feasibility:
This study is carried out to check the technical feasibility, that is, the technical
requirements of the system. Any system eveloped must not have a high demand on the available
technical resources. This will lead to high demands on the available technical resources. This
will lead to high demands being placed on the client. The developed system must have a modest
requirement, asonly minimalor null changes are required for implementing this system.

3.1.3 Social feasibility:

The aspect of study is to check the level of acceptance of the system by the user. This
includes the process of training the user to use the system efficiently. The user must not feel
threatened by the system, instead must accept it as a necessity. The level of acceptance by the
users solely depends on the methods that are employed to educate the user about the system
and to make him familiar with it. His level of confidence must be raised so that he is also able
to make some constructive criticism, which is welcomed, as he is the final user ofthe system.

Fraud Prevention In Real Time Mobile Social Networks Page No:6

 4. SYSTEM ANALYSIS

4.1 EXISTING SYSTEM:

• In recent years, research that explores social relationship structure for information diffusion
in MSNs has been very active. Especially, the problem of maximizing the influence of
information has attracted the attention from both the academia and industry, and a number
of innovative research results.
• At present, the research on information diffusion mainly develops along two branches:

1.Modeling ofthe information diffusion process

2.Controlof information diffusion process.

• In view of the modeling of the information diffusion process, most scholars use the
infectious disease diffusion model, the independent cascade model, the linear threshold
model, the real dataset fitting method, and so on, to model the spatio-temporal dynamic
evolutionaryprocess of information diffusion.

4.2 DISADVANTAGES OF EXISTING SYSTEM:

• System is less effective due to lack of thinking, trust, and diffusion, the three psychological
cognitive and behavioral states.
• The system doesn’t effective since gradually lose the awareness of fraud information due
to its forgetting psychology, it may be infected again by fraud information in the future.

4.3 PROPOSED SYESTEM:

In the proposed system, the system put forward a novel dynamics model, called SWIR, which
can accurately describe the dynamic process of fraud information diffusion. Importantly, for
the sake of efficiently utilizing the limited resources and minimizing the losses of individuals,
we establish the optimal control system to solve the optimal dynamic allocation problem of
control strategies for fraud information diffusion. The main contributions of this paper are
summarized as follows.
1. Fraud Information Diffusion Model: In consideration of the uncertain mental state
of individuals and the transitional relationship of individuals in different states, we establish
the SWIR model. It can more effectively describe the dynamic diffusion process of fraud
information in MSNs. In addition, we theoretically analyze the stability of the SWIR model
and the trend of fraud information diffusion.

Fraud Prevention In Real Time Mobile Social Networks Page No:7

 2. Dynamic Allocation of the Control Strategies : In order to efficiently utilize
limited control resources and minimize losses of individuals caused by fraud information, we
propose two synergistic control strategies. We take the control resources consumption and the
losses of individuals as the total cost constraint. Then, we formulate the optimal control
problem to minimize the total cost, and model the control strategies as functions varying over
time. Finally, based on the optimal control theory, the optimal distribution of the control
strategies functions over time is derived.

4.4 ADVANTAGES OF PROPOSED SYSTEM:

• The proposed system establishes an information diffusion model to accurately describe the
dynamic diffusion process of fraud information in MSNs by considering the uncertain
mental states of individuals.
• The system analyzes the trend of information diffusion and the stability of the dynamics
model from a theoretical point of view and explores the theory of dynamic evolution of
information diffusion model.

Fraud Prevention In Real Time Mobile Social Networks Page No:8

 5. SYSTEM REQUIREMENTS

5.1 Hardware Requirements:

 Processor - I3 (Min)

 RAM - 4 GB (min)

 Hard Disk - 120 GB (Min)

5.2 Software Requirements:

 Operating System - Windows10

 Technology Used - HTML

Fraud Prevention In Real Time Mobile Social Networks Page No:9

 6. SYSTEM DESIGN

6.1 System Architecture:

Fraud Prevention in Real Time Mobile social Networks Page No:10

 6.1.1 Data flow diagram:

1. The term "bubble chart" or "DFD" is also used to describe it. It is possible to visualize
a system in terms of data intake, processing, and output by using a simple visual
formalism.
2. Data flow diagram (DFD) is a vital modeling tool. 2. the system's parts are what this
tool is utilized for This list includes system procedures, data, an external entity, and
information flow within the system.
3. A data flow diagram can be seen in three different ways (DFD). For example, it
shows how input and output are transformed in a visual representation of this
process.
4. DFD, or bubble chart, is a type of graph. A DFD can be used at any level of
abstraction to depict a system. DFD can be broken down into a series of levels that
indicate increasing levels of information flow and operation.

Fraud Prevention In Real Time Mobile Social Networks Page No:11

[T y pe h er e]
 6.1.2 Data owner

Fraud Prevention In Real Time Mobile Social Networks Page No:12

6.1.3 Data user

 Flow Chart : User

Start

User Register

Yes No
Login

View users Profile Username& Password

Wrong

List All Users and Follow

List All Follow Request

View All My Friends Logout

Upload Post

View All My Posts,View All

Friends Posts.

Fraud prevention In real Time Mobile Social Networks Page No:13

 6.2 MODULES DESCRIPTION

Admin:
In this module, the Admin has to login by using valid user name and password. After
login successful he can perform some operations such as View All Users And Authorize,
View All Friends Details, Add Filter, View All Posts, View All Reviews, View Fraud Info
Spreading, View Likes Results.
Friend Request and Response:

In this module, the admin can view all the friend requests and responses. Here all the
requests and responses will be displayed with their tags such as Id, requested user photo,
requested username, user name request to, status and time & date. If the user accepts the
request then the status will be changed to accepted or else the status will remains as waiting.
User:
In this module, there are n numbers of users are present. User should register before
performing 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.
Verify finger print and Login Once Login is successful user can perform some operations like
List All Users and Follow, List All Follow Request, View All My Friends, Upload Post, View
All My Posts, View All Friends Posts.
Searching Users to make friends:
In this module, the user searches for users in Same Network and in the Networks and
sends friend requests to them. The user can search for users in other Networks to make
friends only if they have permission.

Fraud prevention In real Time Mobile Social Networks Page No:14

 6.3 UML DIAGRAMS
6.3.1 USE CASE DIAGRAM:
Diagrams in the Unified Modeling Language (UML) based on use-case studies are
known as use-case diagrams. This graphic was intended to show how a system works in
terms of actors, goals (represented as use cases), and any interdependencies among those use
cases. It's possible to indicate which actors are accountable for carrying out system functions
using a use case diagram (UCD). Actors may be seen in great detail.

Fraud Prevention In Real Time Mobile Social Networks Page No:15

 6.3.2 CLASS DIAGRAM:
When it comes to software engineering, a class diagram is a type of static structural
diagram that depicts all the classes, attributes, operations (or methods), and relationships
between those classes in the Unified Modeling Language (UML). There is a description for
the type of data that is held there.

Fraud Prevention In Real Time Mobile Social Networks Page No:16

 6.3.3 SEQUENCE DIAGRAM:
Using the Unified Modeling Language, sequence diagrams are diagrams that explain
how processes interact and in what order they do so (UML). What we have here is MSC
construct. Depending on the context, sequence diagrams might be referred to as event
diagrams, events scenarios or timing diagram.

Fraud Prevention In Real Time Mobile Social Networks Page No:17

 7.SYSTEM IMPLEMENTATION
7. 1 Software Environment:
The Java Language When it comes to Java, you're getting a language and a platform in one.Java,
one of the most extensively used programming languages on the planet, is the programming
language Java is described by the following buzzwords:

To put it simply,In terms of architectural style, it is neutral.Object-oriented design Conveniently

sized A Dispersed’ Incredibly efficient Translated In a multi-threaded environment Stable
‘Dynamic’ is a synonym for 'dynamic' Ensured Compiling or interpreting a program written in one
of the many popular programming languages allows you to run it on your computer. Java is an
unusual programming language since it can be both compiled and interpreted. The compiler
generates Java byte codes from platform-independent code, which is then interpretedby the Java
platform's interpreter. The interpreter parses and executes each Java bytecode instruction on the
computer. While compilation occurs only once, each time aprogram is run, interpretation occurs on
every run. The graphic below illustrates this.

7.1.1 General J2ME architecture

Configurations and profiles are used by J2ME to customize the Java Runtime Environment (JRE).
With the addition of domain-specific classes and a Java Runtime Environment (JRE), J2ME is a full
Java Runtime Environment (JRE).This means that only certain types of hardware can run a specified

Fraud Prevention In Real Time Mobile Social Networks Page No:18

 set of core classes anda specific JVM. The course will go into great detail on how to set up various
components. To put it another way, this profile adds specific domain-specific classesto the J2ME
setup to identify specific uses for devices. The many virtual machines,settings, and profiles are
depicted in the diagram below. We'll go into great detail onprofiles. Java virtual machine and the
J2SE API share many similarities. It is customary to refer to the Java Virtual Machine (JVM) as a
whole when discussing the J2ME virtual machines (KVM and CVM). Although they are shortened
versionsof the J2SE JVM, the J2ME-specific KVM and CVM can be considered Java virtual
machines.

7.1.1. Building J2ME apps is the next stage

Introduction creating apps for mobile devices of all sizes has a few considerations that will be
explored in this section. If you're using J2SE to compile J2ME apps, look at how the compiler is
called. The packaging and deployment procedure wouldbe incomplete without pre verification.

In the design of compact devices, there are three factors to consider.When creating software for
mobile devices, it's important to keep a few things in mind. It's a good idea to go through your use
case before diving into the code for a small device. It can be a frustrating experience to have to redo
the code since you failed to account for all of the "gotchas".

Take a look at the following options:

Don't complicate things. You can either remove unnecessary functionality or createa separate app
for them.

It is advisable to have a smaller size. This should be taken into consideration by alldevelopers.
Because they are smaller, apps can be downloaded and installed more rapidly. When it comes to
distributing your apps, think about using Java Archive (jar) files.

Keep RAM usage to a minimum while the app is running. The amount of memory required at run
time can be reduced by using scalar rather than object types. Don't trust the trash hauler, either.
Memory can be conserved by converting unused objectreferences to null. Allowing objects to be
allocated only when they are needed decreases the amount of memory required at runtime. In order
to limit memory consumption on small devices, it is important to release resources as soon as
possible, reuse objects, and avoid exceptions.

a list of the many configuration,This means that only certain types of hardware can run a specified
set of core classesand a specific JVM. There are currently just two J2ME configurations available:

In order to use the KVM's Connected Limited Device Configuration, you must use 16-bit or 32-bit
devices with a small amount of memory (CLDC). This configuration is commonly used for
developing small J2ME apps (along with the corresponding virtual machine). CLDC is more

Fraud Prevention In Real Time Mobile Social Networks Page No:19

 difficult to develop than CDC due to its smaller size. CLDC is the setting we'll be using for our
drawing tool application. A small wireless device running a small application is an example of a
Palm hand-held computer.

Configured Device Configuration (CVC) is used by the CVM on 32-bit systems requiring more than
2 MB of memory (CDC). Devices like the Net TV box fall under this category.

J2ME profiles number five

In this session, we spoke about the importance of a profile in determining which devices are
supported. The Mobile Information Device Profile (MIDP) establishes the class of a mobile phone
(MIDP). It adds domain-specific classes to the J2ME configuration in order to find usage for related
devices. KJava and MIDP are the twoCLDC-based J2ME profiles. Both KJava and MIDP are
connected with CLDC andsmaller devices. Profiles are built on configurations. Memory (memory
capacity) ofthe device on which an application is running is linked to certain profile settings.

There is a Foundation Profile from which you can design your own, a skeleton profile.

kJava is the first profile in this list.

The KJava API is contained in a Sun profile named KJava, which is Sun's proprietary profile. The KJava
profile is constructed from the CLDC setup. The KVM virtual machine accepts the same byte codes
and class file formats as the J2SE virtual machine. KJava includes the Sun-specific API, which runs
on Palm OS. Both the J2SE AWT and KJava API have a lot of similarities (AWT). So its default
J2ME package has been replaced by com.sun.kjava. A better understanding of the KJava API will
be gained after writing some sample programs.
7.1.2. In this section, you will find the MIDP profile:
MIDP is primarily aimed for mobile devices like smartphones and pagers. It is possible to
dynamically install new applications and services on end-user devices thanks to the MIDP, which
is based on CLDC, like KJava. Because it is not specificto any one manufacturer, MIDP has become
a mobile device industry standard. It's an all-in-one development platform for mobile apps,
complete with comprehensivedocumentation and assistance. It's worth noting that only the first
three of the following packages are CLDC-specific.

The java.lang package containsIo * in the JDK

Interfaces in the java.util
The Java Micro Edition Interface is implemented by this class (JMI)
Interface for Microedition Objects in Java (LCDUI)
* javax.microedition.microedition
The Java Micro Edition Reference Model is implemented in this class.

Fraud Prevention In Real Time Mobile Social Networks Page No:20

 7.2.SOURCE CODE

Fraud Prevention In Real Time Mobile Social Networks Page No:21

Fraud Prevention In Real time Mobile social Networks page No:22
Fraud Prevention Real Time Mobile Social Networks Page No:23
Fraud Prevention In Real Time Mobile Social Networks Page No:24
Fraud Prevention In Real Time Mobile Social Networks Page No:25
Fraud Prevention In Real Time Mobile Social Networks Page No:26
Fraud Prevention In Real Time Mobile Social Networks Page No:27
Fraud Prevention In Real Time Mobile Social Networks Page No:28
Fraud Prevention In Real Time Mobile Social Networks Page No:29
Fraud Prevention In Real Time Mobile Social Networks Page No:30
Fraud Prevention In Real Time Mobile Social Networks Page No:31
 8. SYSTEM TESTING

8.1 Testing
The eighth step is system testing.
Testing is a must in order to discover errors. Testing is the practice of looking for any and all flaws
or vulnerabilities in a product. This method can be used to verify the performance of components,`
subassemblies, assemblies, and finished products. Testing software to see if it does what it's
supposed to do.
The software system's specs and user expectations have no undesirable faults. Thereis a wide range
of testing options. For each kind of test, there's a certain purpose forwhich it was created.
8.1.1 The methods of assessment
Each unit must be tested individually.

If you want to make sure your software works as expected and that your inputs andoutputs are
valid, you utilize unit testing. Ensure that all code paths and decision points are tested thoroughly.
An application's software components are tested separately.. Each individual unit must be
completed before integration may take place. These are structural tests, which demand an intrusive
process and require knowledge of its construction. Unit tests are conducted at the component level
to check a specific business process, application, or system configuration. Unit tests are one method
of verifying that a business process adheres to its documented criteria.

8.1.2 Implementation and evaluation of results

If two or more pieces of software can actually work together as a single entity, an integration test
is needed. The outcomes of screens and fields are more important inan event-driven testing
methodology. It is important to do integration tests in order to make sure that all of the components
work together in a logical and consistent manner. Integration testing is used to detect the difficulties
that arise when two or more components are combined.

8.1.3 A test of utility

It is important to conduct functional tests to ensure that all of the requirements of thebusiness and
technical systems as well as system documentation and user guides have been met.

Functional testing focuses on the following:

It is essential to only accept authentic input that has been specifically identified.
Identification and rejection of specific types of incorrect input.
In order for the system to function correctly, it is necessary to complete several tasks.
For each of the application output classes, outputs must be tested.

Fraud Prevention In Real Time Mobile Social Networks Page No:32

 Systems and methods for interfacing with other systems and procedures must be established.

Functional tests are organized and prepared based on requirements, significant functions, or
specialized test cases. Additional testing is required to discover business processes, data fields and
established procedures. Additional testing needsto be discovered and the value of current tests
established in order to complete functional testing, An evaluation of the 8.1.4 version of the system.
In order to ensure that all of the requirements are met, the entire integrated softwaresystem is tested.
A test's outcome can now be predicted in advance as a result of this.A good example of how a system
can be tested is the configuration-oriented systemintegration test. Using process descriptions and
flow maps, system testing emphasizes pre-driven links and integration points.
In 8.1.5, white box testing
"White box testing" refers to a type of testing in which the software tester knows theproduct's inner
workings, structure, language, and programming language. It has a purpose. In order to examine
places where a regular black box can't reach, this deviceis employed.
8.2 Black Box Testing
It is the process of testing software without having any understanding of the modulebeing tested, its
underlying structure, language, or design. For black box testing to be effective it must come from
a clear source document, such as a specification or requirements document. The term "black box
testing" refers to software testing in which the program under test is treated as a mystery. One
cannot "see" inside. Testsare conducted only on input and outputs, not on the underlying software.

When it comes to the 8.2 unit tests,

During the software development process, unit testing and coding may be accomplished at the same
time. It is also possible to segregate unit testing from othertesting.

It is imperative that strategies and tactics be put to the test.

Detailed functional tests will be written up for future reference during field testing.
The test's aims and objectives
There should be no mistakes in the data.

Only by clicking on the link supplied will you be able to view the requested pages.
Sending a message or entering data shouldn't necessitate any sort of wait time onyour end.

The features will be put through their paces in testing.

Make that the data is in the correct format by performing an audit on it.
No more than one entry should be allowed in the database.
Every link needs to point to the appropriate page.
8.3 Integrity testing of the various parts

Fraud Prevention In Real Time Mobile Social Networks Page No:33

 Softwareintegration testing refers to the process of integrating software componentsone at a time to
look for interface issues that could lead to problems.

In an integration test, the primary purpose is to ensure that all of the components ofa system or
company-wide software application interact with each other.

Everysingle one of the above-mentioned tests went perfectly. Everything worked asit should.

8.4 This is the 8.4th test.

End-User Testing is a crucial part of any project, and the end user must play a significant role. The
system is further checked to ensure that it meets all necessary functional criteria.
Every single one of the above-mentioned tests went perfectly. None of the concerns were
discovered.

Fraud Prevention In Real Time Mobile Social Networks Page No:34

 9. RESULTS

Fig:9.1 Starting home page

Fig:9.2 Admin login page

Fraud Prevention In Real Time Mobile Social Networks Page No:35

 Fig:9.3 Admin page view

Fig:9.4 Add filter details and existing filter details

Fraud Prevention In Real Time Mobile Social Networks Page No:36

 Fig:9.5 User login page

Fig:9.6 User registration form

Fraud Prevention In Real Time Mobile Social Networks Page No:37

 Fig:9.7 User page view

Fig:9.8 Post uploading process

Fraud Prevention In Real Time Mobile Social Networks Page No:38

 10. CONCLUSION
The goal of this project is to put forward the optimal control strategies to efficiently
utilize limited control resources and minimize losses of individuals caused by the diffusion of fraud
information. First, a novel SWIR dynamics model is proposed to describe the dynamic evolutionary
process of fraud information diffusion in MSNs. Thereafter, this paper analyzes and proves the
information diffusion trends and stability of the dynamics model. In particular, this paper proposes
two synergistic control strategies to suppress the spread of fraud information, and derives the
optimal dynamic allocation of the control strategies. Finally, we validate the efficiency of our
proposed diffusion model and optimal control strategies in both synthetic datasets and real social
network datasets. This paper can provide a theoretical basis and a feasible technical approach for
the applications of controllable information diffusion based on MSNs, and further promote the
development and application of information diffusion and optimal control technology in MSNs. In
the future, we will further study the diffusion modeling and control of coupling of positive and
negative information. In addition, we will also study the impact of users’ social identity cognition
on information diffusion.

Fraud Prevention In Real Time Mobile Social Networks Page No:39

 11. REFERENCES

[1] M. Xiao, J. Wu, L. Huang, R. Cheng, and Y. Wang, “Online task assignment for crowdsensing
in predictable mobile social networks,” IEEE Trans. Mobile Comput., vol. 16, no. 8, pp. 2306–
2320, Aug. 2017.

[2] L. Jiang, J. Liu, D. Zhou, Q. Zhou, X. Yang, and G. Yu, “Predicting the evolution of hot topics:
A solution based on the online opinion dynamics model in social network,” IEEE Trans. Syst.,
Man, Cybern., Syst., to be published.

[3] Y. Lin et al., “An on-demand coverage based self-deployment algorithm for big data perception
in mobile sensing networks,” Future Gener. Comput. Syst., vol. 82, pp. 220–234, May 2018.

[4] Y. Wang, A. V. Vasilakos, J. Ma, and N. Xiong, “On studying the impact of uncertainty on
behavior diffusion in social networks,” IEEE Trans. Syst., Man, Cybern., Syst., vol. 45, no. 2, pp.
185–197, Feb. 2015.

[5] L.-X. Yang, P. Li, Y. Zhang, X. Yang, Y. Xiang, and W. Zhou, “Effective repair strategy against
advanced persistent threat: A differential game approach,” IEEE Trans. Inf. Forensics Security, vol.
14, no. 7, pp. 1713–1728, Jul. 2019.

[6] Z. He, Z. Cai, J. Yu, X. Wang, Y. Sun, and Y. Li, “Cost-efficient strategies for restraining rumor
spreading in mobile social networks,” IEEE Trans. Veh. Technol., vol. 66, no. 3, pp. 2789–2800,
Mar. 2017.

[7] L.-X. Yang, P. Li, X. Yang, Y. Wu, and Y. Y. Tang, “On the competition of two conflicting
messages,” Nonlin. Dyn., vol. 91, no. 3, pp. 1853–1869, 2018.

[8] R. Nash, M. Bouchard, and A. Malm, “Investing in people: The role of social networks in the
diffusion of a large-scale fraud,” Soc. Netw., vol. 35, no. 4, pp. 686–698, 2013.

[9] R. A. Raub, A. H. N. Hamzah, M. D. Jaafar, and K. N. Baharim, “Using subscriber usage profile
risk score to improve accuracy of telecommunication fraud detection,” in Proc. IEEE
CYBERNETICSCOM, 2016, pp. 127–131.

Fraud Prevention In Real Time Mobile Social Networks Page No:40

 [10] J. Ma et al., “Detecting rumors from microblogs with recurrent neural networks,” in Proc.
IJCAI, 2016, pp. 3818–3824.

[11] (Aug. 2016). Tsinghua University Teachers Cheated 17 Million 600Thousand? The Original
Liar Used This Psychological Routine! [Online].
Available: http://www.bestchinanews.com/Domestic/2426.html

[12] M. Sahin, “Over-the-top bypass: Study of a recent telephony fraud,” in Proc. ACM CCS, 2016,
pp. 1106–1117.

[13] K. Zhu and L. Ying, “Information source detection in the SIR model: A sample-path-based
approach,” IEEE/ACM Trans. Netw., vol. 24, no. 1, pp. 408–421, Feb. 2016.

[14] Z. Chen, K. Zhu, and L. Ying, “Detecting multiple information sources in networks under the
SIR model,” IEEE Trans. Netw. Sci. Eng., vol. 3, no. 1, pp. 17–31, Jan./Mar. 2016.

[15] A. Y. Khrennikov, Information Dynamics in Cognitive, Psychological, Social, and Anomalous

Phenomena, vol. 138. New York, NY, USA:Springer, 2013.

[16] R. Lachman, J. L. Lachman, and E. C. Butterfield, Cognitive Psychology and Information

Processing: An Introduction. London, U.K.:Psychology, 2015.

[17] S. Wen, M. S. Haghighi, C. Chen, Y. Xiang, W. Zhou, and W. Jia, “A sword with two edges:
Propagation studies on both positive and negative information in online social networks,” IEEE
Trans. Comput., vol. 64, no. 3, pp. 640–653, Mar. 2015.

[18] E. Kušen, M. Strembeck, G. Cascavilla, and M. Conti, “On the influence of emotional valence
shifts on the spread of information in social networks,” in Proc. IEEE/ACM ASONAM, 2017, pp.
321–324.

[19] K. Kandhway and J. Kuri, “Using node centrality and optimal control to maximize information
diffusion in social networks,” IEEE Trans. Syst., Man, Cybern., Syst., vol. 47, no. 7, pp. 1099–
1110, Jul. 2017.

[20] A. Nematzadeh, E. Ferrara, A. Flammini, and Y.-Y. Ahn, “Optimal network modularity for
information diffusion,” Phys. Rev. Lett., vol. 113, no. 8, 2014, Art. no. 088701.

Fraud Prevention In Real Time Mobile Social Networks Page No:41

 [21] K. Kandhway and J. Kuri, “How to run a campaign: Optimal control of SIS and SIR
information epidemics,” Appl. Math. Comput., vol. 231, no. 1, pp. 79–92, 2014.

[22] X. Wang, Y. Lin, Y. Zhao, L. Zhang, J. Liang, and Z. Cai, “A novel approach for inhibiting
misinformation propagation in human mobile opportunistic networks,” Peer-to-Peer Netw. Appl.,
vol. 10, no. 2, pp. 377–394, 2017.

[23] Q. Zhao, C. Wang, P. Wang, M. Zhou, and C. Jiang, “A novel method on information
recommendation via hybrid similarity,” IEEE Trans. Syst., Man, Cybern., Syst., vol. 48, no. 3, pp.
448–459, Mar. 2018.

[24] Y. Jiang and J. C. Jiang, “Diffusion in social networks: A multiagent perspective,” IEEE Trans.
Syst., Man, Cybern., Syst., vol. 45, no. 2, pp. 198–213, Feb. 2015.

[25] L.-X. Yang, X. Yang, and Y. Y. Tang, “A bi-virus competing spreading model with generic
infection rates,” IEEE Trans. Netw. Sci. Eng., vol. 5, no. 1, pp. 2–13, Jan./Mar. 2018.

Fraud Prevention In Real Time Mobile Social Networks Page No:42

The International journal of analytical and experimental modal analysis ISSN NO: 0886-9367

Fraud Prevention In Real Time Mobile Social

Networks
P. Amulya M Sivani
D Gayathri
Student, Dept. of CSE Student, Dept. of CSE
Student, Dept. of CSE
QIS Institute of Technology QIS Institute of Technology
QIS Institute of Technology
Ongole-523001, India Ongole-523001, India
Ongole-523001, India
amulyapara@gmail.com
gayathriyellamanda@gmail.com maramsivani2001@gmail.com

Y V L Swetha A Vennela
Student, Dept. of CSE Asst. Professor, Dept. of CSE
QIS Institute of Technology QIS Institute of Technology
Ongole-523001, India Ongole-523001, India
yeluriswetha47@gmail.com vennela.alladi@gmail.com

Abstract—Through the use of mobile devices, people can access
real-time information services provided by mobile social
networks (MSNs). MSNs have significant levels of transparency
and autonomy, yet they have been plagued by widespread
misinformation, fraud, and other forms of abuse as a result.
Limiting the availability of fraud alerts is crucial for preventing
further damage. How can scarce resources be best put to use
while still allowing individualsto suffer as little financial loss as
possible as a result of fraudulent data? As a result, we formulate
the challenge of controlling fraudulent data as an optimal
control problem, with the overall cost being the sum of the
control resources expended in enacting control techniques and
the losses incurred by people. We develop the best dynamic
allocation of control strategies based on the idea of optimum
control. Additionally, we build a dynamics model for fraud
information diffusion by factoring in individuals'
unpredictability of mind, and we explore the pattern of fraud
information diffusion and the stability of the dynamics model.
Keywords—Mobile social network, Fraud detection, dynamics
model, fradulent data.
I. INTRODUCTION
With the boom of the Internet and the rapid popularization
of intelligent mobile devices, mobile social networks
(MSNs) have grown up to become an important platform for
information dissemination [1]. MSNs can provide people with
a variety of real-time information services and have already
penetrated into our daily life. The Internet-based MSNs
have exhibited their great charm and broad prospect in
many application fields, such as instant communication, life
service, interactive entertainment, etc., and have attracted
extensive attention of the industry and the academia [2], [3].
However, the development of MSNs is like a double-edged
sword [4], [5]. When MSNs are increasingly becoming an
indispensable part of people’s lives, a series of unhealthy
phenomena, such as fake news, rumors, online promotion,
and fraudulent activities are becoming more and more
rampant, which pose a serious threat on the normal social
network activities [6], [7]. Besides, by means of the
emerging technologies of intelligent terminals, wireless
networks, and online payment in recent years, the high rate
of fraud has caused great losses to people [8]. According to
the official data released by the security ministry,
telecommunications fraud in MSNs has grownat an annual
rate of 20%–30% [9]. The following are two representative
scenarios
Scenario A: One scenario is the Veracruz incident in August
2015 [10].A piece of rumor saying ―shootouts and
kidnappings by drug gangs happening near schools in
Veracruz‖ spread in Twitter and Face book. This rumor
caused severe chaos in the city and many seriouscar crashes
happened amid the hysteria.
Scenario B: Another shocking scenario occurred in August
2016 when a Chinese university professor suffered a
telecommunication-based fraud, lead into a serious loss of
17.6 million Yuan [11]. Criminals fabricated an elaborate
hoax, used the network to transmit fraud information and
perform remote fraud victims.
Fraud information diffusion has become a prominent
problem in social networks [12]. Those evidence highlight
that effectively controlling the fraud information in MSNs
applications is of great significance. Here, we define so-
called fraud information as a piece ofmalicious information
or false information, which aims to intentionally cause
adverse effects, such as mass panic or defraud victims of
their property. In order to cope up with the spread ofsuch
information in MSNs more effectively, it is an urgent need
to study the pattern of fraud information diffusion and further
put forward the corresponding control measures.
Previously, some mathematical models have been used to
model the diffusion evolutionary process of fraud information
in the network. Most of these models are based on the theory
of biological infectious disease because the spread process of
infectious diseases inbiology and the diffusion process of fraud

Volume XV, Issue III, March/ 2023 Page No: 7

The International journal of analytical and experimental modal analysis ISSN NO: 0886-9367

information in the network are very similar [13] . The most resources, they ignore the harm of fraud information diffusion to
widely used model is the susceptible-infected recovered (SIR) individuals.
model, in which all individuals are divided into three
In order to overcome the above limitations, in this paper, we
categories: 1) susceptible; 2) infected; and 3) recovered [14].
put forward a novel dynamics model, called SWIR, which can
From the perspective of information diffusion, the semantics
accurately describe the dynamic process of fraud information
of susceptible, infected, and recovered can fully correspond to
diffusion. Importantly, for the sake of efficiently utilizing the
the process of fraud information diffusion. Ifan individual has
limited resources andminimizing the losses of individuals, we
not yet received any fraud information, it belongs to the
establish the optimal control system to solve the optimal
susceptible state. If an individual received fraud information
dynamic allocation problem of control strategies for fraud
and was misled, it belongs to the infected state. Ifanindividual
information diffusion. The main contributions of this paper are
was ever infected and now no longer believes the fraud
summarized as follows.
information, it belongs totherecovered state.
Fraud Information Diffusion Model: In consideration of
Although the existing SIR-based derivation models can
the uncertain mental state of individuals and the transitional
correctly describe thetransitional relationship and the dynamic
relationship of individuals in different states, we establish the
evolutionary processes of node states, the spread of fraud
SWIR model. It can more effectively describe the dynamic
information in MSNs shows some new characteristics. First,
diffusion process of fraud information in MSNs. In addition,
the information sender and receiver are human beings, and
we theoretically analyze the stability of the SWIR model andthe
human mental activities are often complex. For example, the
trend of fraud information diffusion.
individual will likely experience a series of mental activities,
Dynamic Allocation of the Control Strategies: In order
such as thinking, hesitating, and wandering when receiving a
to efficiently utilize limited control resources and minimize
piece of information [15], [16]. Second, the fraud information
losses of individuals caused by fraud information, we propose
diffusion processes in MSNs are the complex results of the
to synergistic control strategies. We take the control resources
continuous interactions of nodes in different states [17]. Third,
consumption and the losses of individuals as the total cost
because of the psychological effect, repeated reception of the
constraint. Then, we formulate the optimal control problem to
same information may give users the feeling of disgust and
minimize the total cost, and model the control strategies as
lead to reverse psychology. The data analysis about
functions varying over time. Finally, based on the optimal
4.4 million Twitter messages diffusion shows that in the control theory, the optimal distribution of the control strategies
process of information diffusion, users will deviate from the functions over time is derived.
original intention of information and produce the phenomenon Simulation Experiments on Datasets: We validate the
ofemotional transfer [18]. correctness and efficiency of the proposed diffusion model and
the optimal control strategies on both synthetic datasets and
Besides establishing dynamics models and revealing fraud
real social network datasets. The results demonstrate that our
information diffusion laws, our ultimate goal should be to
proposed diffusion model can accurately describe the dynamic
effectively control the diffusion of fraud information.
diffusion process of fraud information and our proposed
However, the implementation of any control or intervention for
control strategies can effectively inhibit the fraud information
the system will incur a certain ―price‖ [19]. As for the process
in MSNs. In particular, the optimal dynamic allocation control
of controlling fraud information diffusion in MSNs, some
strategies can achieve minimum control resources
operational control measures will inevitably consume a certain
consumption and losses of individuals.
amount of precious manpower and material resources. For
The rest of this paper is organized as follows. In Section II,
example, in response to the crisis of fraud information, the
some previous works are reviewed. In Section III, we first
government constantly sends authoritative messages to the
establish a novel dynamics model of the fraud
network to prevent individuals from being misled by it. All
informationdiffusion in MSNs. Then, we analyze the trend of
this need to cost a lot of limited communication and other
fraud information diffusion and the stability of the dynamics
resources. Therefore, how to efficiently utilize limited control
model. Consequently, we propose two synergistic control
resources and minimize losses of individuals by adopting
strategies to suppress the spread of fraud information and
proper control strategies have become an urgent issue to
derive the optimal distribution of the control strategies. The
address.
extensive simulations are conducted in Section IV. Section V
Some of the existing research works can control the diffusion concludes this project.
of fraud information to some extent, but there are still some
obvious issues [20],[21],[22]. The first issue that the they II. RELATED WORKS
usually adopts a single continuous or pulse control strategy, (1) Online task assignment for crowdsensing in predictable
and mostly do not consider the implementation efficiency of mobile socialnetworks
the control strategy and the utilization efficiency of the control. Authors:M. Xiao, J. Wu, L. Huang, R. Cheng, and Y. Wang
The second issue is that while some works have realized the
constraint of control resources and transformed the control Mobile crowdsensing is a new paradigm in which a crowd of
problem into the optimal dynamic allocation of control mobile users exploit their carried smart phones to conduct
complex sensing tasks. In this paper, we focus on the make

Volume XV, Issue III, March/ 2023 Page No: 8

The International journal of analytical and experimental modal analysis
span sensitive task assignment problems for the crowd sensing
in mobile social networks, where the mobility model is
predicable, and the time of sending tasks and recyclingresults is
non-negligible. To solve the problems, we propose an Average
make span sensitive Online Task Assignment (AOTA)
algorithm and a Largest make span sensitive Online Task
Assignment (LOTA) algorithm. In AOTA and LOTA, the
online task assignments are viewed as multiple rounds of
virtual offline task assignments. Moreover, a greedy strategy
of small-task-first-assignment and earliest-idle- user-receive-
task is adopted for each round of virtual offline task
assignment in AOTA, while the greedy strategy of large-task-
first- assignment and earliest- idle-user-receive-task is adopted
for the virtual offline task assignments in LOTA. Based on the
two greedy strategies, both AOTA and LOTA can achieve
nearly optimal online decision performances. We prove this and
give the competitive ratios of the two algorithms. In addition,
we also demonstrate the significant performance of the two
algorithms through extensive simulations, based on four real
MSN traces and a synthetic MSN trace.
Predicting the evolution of hot topics: A solution based on
the online opinion dynamics model in social network
Authors: L. Jiang, J. Liu, D. Zhou, Q. Zhou, X. Yang,
and G. Yu. Predicting and utilizing the evolution trend of hot
topics is critical for contingency management and decision-
making purposes of government bodies and enterprises. This
paper proposes a model named online opinion dynamics
(OODs) where any node in a social network has its unique
confidence threshold and influence radius. The nodes in the
OOD are mainly affected by their neighbors and are also
randomly influenced by unfamiliar nodes. In the traditional
opinion model, however, each node is affected by all other
nodes, including its friends. Furthermore, many traditional
opinion evolution approaches are reviewed to see if all nodes
(participants) can eventually reach a consensus. On the
contrary, OOD is more focused on such details as concluding
the overall trend of events and evaluating the support level of
each participant through numerical simulation. Experiments
show that OOD is superior to the improvement of the original
Hegselmann-Krause (HK) model, HK-13 and HK-17, with
respect to qualitative predictions of the evolution trend of an
event. The quantitative predictions of the HK model cannot be
used to make decisions, whereas the results of the OOD model
are proved to be acceptable. An on-demand coverage based
self-deployment algorithm for big data perception in
mobile sensing networks
Authors: Y. Lin et al
Mobile Sensing Networks have been widely applied to
many fields for big data perception such as intelligent
transportation, medical health and environment sensing.
However, in some complex environments and unreachable
regions of inconvenience for human, the establishment of the
mobile sensing networks, the layout of the nodes and the
control of the network topology to achieve high performance
sensing of big data areincreasingly becoming a main issue in
the applications of the mobile sensing networks. To deal with
Volume XV, Issue III, March/ 2023
ISSN NO: 0886-9367
this problem, we propose a novel on-demand coverage based
self- deployment algorithm for big data perception based on
mobile sensing networking this project. Firstly, by considering
characteristics of mobile sensing nodes, we extend the cellular
automata model and propose a new mobile cellular automata
model for effectively characterizing the spatial–temporal
evolutionary process of nodes. Secondly, based on the
learning automata theory and the historical information of
node movement, we further explore a new mobile cellular
learning automata model, in which nodes can self-adaptively
and intelligently decide the best direction of movement with
low energy consumption. Finally, we propose a new
optimization algorithm which can quickly solve the node self-
adaptive deployment problem, thus, we derive the best
deployment scheme of nodes in a short time. The extensive
simulation results show that the proposed algorithm in this
paper outperforms the existing algorithms byas much as 40% in
terms of the degree of satisfaction ofnetwork coverage, the
iterations of thealgorithm, the average moving steps of nodes
and the energy consumption of nodes. Hence, we believe that
our work will make contributions to large-scale adaptive
deployment and high performance sensing scenarios of the
mobile sensing networks. Quickly solve the node self-adaptive
deployment problem, we derive the best deployment scheme of
nodes in a short time.The extensive simulation results show that
the proposed algorithm in this project out performsthe existing
algorithms by as much as 40% in terms of the degree of
satisfaction of network coverage, the iterations of the
algorithm, the average moving steps of nodes and the energy
consumption of nodes. Hence, we believe that our work will
make contributions to large- scale adaptive deployment and
high performance sensing scenarios of the mobile sensing
networks.In recent years, research that explores social
relationship structure for information diffusion in MSNs has
been very active. Especially, the problem of maximizing the
influence of information has attracted the attention from both
the academia and industry, and a number of innovative
research results.At present, the research on information
diffusion mainly develops along two branches: 1.Modeling
ofthe information diffusion process
2.Controlof information diffusion process. In view of the
modeling of the information diffusion process, most scholars
use the infectious disease diffusion model, the independent
cascade model, the linear threshold model, the real dataset
fitting method, and so on, to model the spatio-temporal
dynamic evolutionaryprocess of information diffusion.
DISADVANTAGES OF EXISTING SYSTEM:
System is less effective due to lack of thinking, trust, and
diffusion, the three psychological cognitive and behavioral
states.The system doesn’t effective since gradually lose the
awareness of fraud information due to its forgetting
psychology, it may be infected again by fraud information in
the future.
Page No: 9
The International journal of analytical and experimental modal analysis ISSN NO: 0886-9367

III. PROPOSED SYSTEM user name and password. Verify finger print and Login Once
Login is successful user can perform some operations like List
In the proposed system, the system put forward a novel All Users and Follow, List All Follow Request, View All My
dynamics model, called SWIR, which can accurately describe Friends, Upload Post, View All My Posts, View All Friends
the dynamic process of fraud information diffusion. Posts.
Importantly, for the sake of efficiently utilizing the limited
resources and minimizing the losses of individuals, we 4) Searching Users to make friends:
establish the optimal control system to solve the optimal
dynamic allocation problem of control strategies for fraud In this module, the user searches for users in Same Network
information diffusion. The main contributions of this paper are and in the Networks and sends friend requests to them. The
summarized as follows. user can search for users in other Networks to make friends
only if they have permission.
Fraud Information Diffusion Model: In consideration of the
uncertain mental state of individuals and the transitional
relationship of individuals in different states, we establish the
SWIR model. It can more effectively describe the dynamic
diffusion process of fraud information in MSNs. In addition,
we theoretically analyze the stability of the SWIR model and
the trend of fraud information diffusion.
Dynamic Allocation of the Control Strategies: In order to
efficiently utilize limited control resources and minimize
losses of individuals caused by fraud information, we propose
two synergistic control strategies. We take the control
resources consumption and thelosses of individuals as the total
cost constraint. Then, we formulate the optimal control
problem to minimize the total cost, and model the control
strategies as functions varying overtime. Finally, based on the
optimal control theory, the optimal distribution of the control
strategies functions over time is derived.
ADVANTAGES OF PROPOSED SYSTEM:
The proposed system establishes an information diffusion
model to accurately describe thedynamic diffusion process of
fraud information in MSNs by considering the uncertain
mental states of individuals.
The system analyzes the trend of information diffusion and the Fig.1 Architecture of proposed system
stability of the dynamics model from a theoretical point of
view and explores the theory of dynamic evolution of
information diffusion model.
1) Admin:
In this module, the Admin has to login by using valid user
name and password. After login successful he can perform
some operations such as View All Users And Authorize, View
All Friends Details, Add Filter, View All Posts, View All
Reviews, View Fraud Info Spreading, View Likes Results.
2) Friend Request and Response:
In this module, the admin can view all the friend requests and
responses. Here all therequests and responses will be displayed
with their tags such as Id, requested user photo, requested
username, user name request to, status and time & date. If the
user accepts the request then the status will be changed to
accepted or else the status will remains as waiting.
Fig.2 DFD for fraud detection
3) User:
In this module, there are n numbers of users are present. User IV. RESULTS AND DISCUSSION
should register before performing any operations. Once user The output screens obtained after running the project code
registers, their details will be stored to the database. After is shown from Fig.3 to Fig.10.
registration successful, he has to login by using authorized

Volume XV, Issue III, March/ 2023 Page No: 10

The International journal of analytical and experimental modal analysis ISSN NO: 0886-9367

Fig.3 Home page Fig.7 User login page

Fig.4 Admin Login Page Fig.8 User registration page

Fig.5 Admin page view

Fig.9 User Page view

Fig.6 Add filter details Fig.10 post uploading process

Volume XV, Issue III, March/ 2023 Page No: 11

The International journal of analytical and experimental modal analysis
V. CONCLUSION
The goal of this project is to put forward the optimal control
strategies to efficiently utilize limited control resources and
minimize losses of individuals caused by the diffusion of
fraud information. First, a novel SWIR dynamics model is
proposed to describe the dynamic evolutionary process of
fraud information diffusion in MSNs. Thereafter, this paper
analyzes and proves the information diffusion trends and
stability of the dynamics model. In particular, this paper
proposes two synergistic control strategies to suppress the
spread of fraud information, and derives the optimal dynamic
allocation of the control strategies. Finally, we validate the
efficiency of our proposed diffusion model and optimal
control strategies in both synthetic datasets and real social
network datasets. This paper can provide a theoretical basis
and a feasible technical approach for the applications of
controllable information diffusion based on MSNs, and
further promote the development and application of
information diffusion and optimal control technology in
MSNs. In the future, we will further study the diffusion
modeling and control of coupling of positive and negative
information. In addition, we will also study the impact of
users’ social identity cognition on information diffusion.
References
[1] M. Xiao, J. Wu, L. Huang, R. Cheng, and Y. Wang, ―Online task
assignment for crowdsensing in predictable mobile social networks,‖
IEEE Trans. Mobile Comput., vol. 16, no. 8, pp. 2306– 2320, Aug.
2017.
[2] L. Jiang, J. Liu, D. Zhou, Q. Zhou, X. Yang, and G. Yu, ―Predicting the
evolution of hot topics:A solution based on the online opinion dynamics
model in social network,‖ IEEE Trans. Syst., Man, Cybern., Syst., to be
published.
[3] Y. Lin et al., ―An on-demand coverage based self-deployment algorithm
for big data perception in mobile sensing networks,‖ Future Gener.
Comput. Syst., vol. 82, pp. 220–234, May 2018.
[4] Y. Wang, A. V. Vasilakos, J. Ma, and N. Xiong, ―On studying the
impact of uncertainty on behavior diffusion in social networks,‖ IEEE
Trans. Syst., Man, Cybern., Syst., vol. 45, no. 2, pp. 185–197, Feb. 2015.
[5] L.-X. Yang, P. Li, Y. Zhang, X. Yang, Y. Xiang, and W. Zhou, ―Effective
repair strategy against advanced persistent threat: A differential game
approach,‖ IEEE Trans. Inf. Forensics Security, vol.14, no. 7, pp. 1713–
1728, Jul. 2019.
[6] Z. He, Z. Cai, J. Yu, X. Wang, Y. Sun, and Y. Li, ―Cost-efficient strategies
for restraining rumorspreading in mobile social networks,‖ IEEE Trans.
Veh. Technol., vol. 66, no. 3, pp. 2789–2800, Mar. 2017.
[7] L.-X. Yang, P. Li, X. Yang, Y. Wu, and Y. Y. Tang, ―On the
competition of two conflicting messages,‖ Nonlin. Dyn., vol. 91, no. 3,
pp. 1853–1869, 2018.
[8] R. Nash, M. Bouchard, and A. Malm, ―Investing in people: The role of
social networks in the diffusion of a large-scale fraud,‖ Soc. Netw., vol.
35, no. 4, pp. 686–698, 2013.
Volume XV, Issue III, March/ 2023
ISSN NO: 0886-9367
[9] R. A. Raub, A. H. N. Hamzah, M. D. Jaafar, and K. N. Baharim, ―Using
subscriber usage profile risk score to improve accuracy of
telecommunication fraud detection,‖ in Proc. IEEE
CYBERNETICSCOM, 2016, pp. 127–131.
[10] J. Ma et al., ―Detecting rumors from microblogs with recurrent neural
networks,‖ in Proc.IJCAI, 2016, pp. 3818–3824.
[11] (Aug. 2016). Tsinghua University Teachers Cheated 17 Million
600Thousand? The OriginalLiar Used This Psychological Routine!
[12] M. Sahin, ―Over-the-top bypass: Study of a recent telephony fraud,‖ in
Proc. ACM CCS, 2016,pp. 1106–1117.
[13] K. Zhu and L. Ying, ―Information source detection in the SIR model: A
sample-path-based approach,‖ IEEE/ACM Trans. Netw., vol. 24, no. 1,
pp. 408–421, Feb. 2016.
[14] Z. Chen, K. Zhu, and L. Ying, ―Detecting multiple information sources
in networks under theSIR model,‖ IEEE Trans. Netw. Sci. Eng., vol. 3,
no. 1, pp. 17–31, Jan./Mar. 2016.
[15] A. Y. Khrennikov, Information Dynamics in Cognitive, Psychological,
Social, and Anomalous Phenomena, vol. 138. New York, NY,
USA:Springer, 2013.
[16] R. Lachman, J. L. Lachman, and E. C. Butterfield, Cognitive
Psychology and Information Processing: An Introduction. London,
U.K.:Psychology, 2015.
[17] S. Wen, M. S. Haghighi, C. Chen, Y. Xiang, W. Zhou, and W. Jia, ―A
sword with two edges:Propagation studies on both positive and negative
information in online social networks,‖ IEEE Trans. Comput., vol. 64,
no. 3, pp. 640–653, Mar. 2015.
[18] E. Kušen, M. Strembeck, G. Cascavilla, and M. Conti, ―On the influence
of emotional valence shifts on the spread of information in social
networks,‖ in Proc. IEEE/ACM ASONAM, 2017, pp.321–324.
[19] K. Kandhway and J. Kuri, ―Using node centrality and optimal control to
maximize informationdiffusion in social networks,‖ IEEE Trans. Syst.,
Man, Cybern., Syst., vol. 47, no. 7, pp. 1099– 1110, Jul. 2017.
[20] A. Nematzadeh, E. Ferrara, A. Flammini, and Y.-Y. Ahn, ―Optimal
network modularity for information diffusion,‖ Phys. Rev. Lett., vol.
113, no. 8, 2014, Art. no. 088701.
[21] K. Kandhway and J. Kuri, ―How to run a campaign: Optimal control of
SIS and SIR information epidemics,‖ Appl. Math. Comput., vol. 231,
no. 1, pp. 79–92, 2014.
[22] X. Wang, Y. Lin, Y. Zhao, L. Zhang, J. Liang, and Z. Cai, ―A novel
approach for inhibiting misinformation propagation in human mobile
opportunistic networks,‖ Peer-to-Peer Netw. Appl., vol. 10, no. 2, pp.
377–394, 2017.
[23] Q. Zhao, C. Wang, P. Wang, M. Zhou, and C. Jiang, ―A novel method
on information recommendation via hybrid similarity,‖ IEEE Trans.
Syst., Man, Cybern., Syst., vol. 48, no. 3, pp.448–459, Mar. 2018.
[24] Y. Jiang and J. C. Jiang, ―Diffusion in social networks: A multiagent
perspective,‖ IEEE Trans. Syst., Man, Cybern., Syst., vol. 45, no. 2, pp.
198–213, Feb. 2015.
[25] L.-X. Yang, X. Yang, and Y. Y. Tang, ―A bi-virus competing spreading
model with generic infection rates,‖ IEEE Trans. Netw. Sci. Eng., vol. 5,
no. 1, pp. 2–13, Jan./Mar. 2018.
Page No: 12