Index

Sr.No Topic Name Page No

1 Acknowledgement 2

2 Abstract 3

3 Aim of the project 3

4 Course Outcome 3

5 Case Study of Fake Profiling 7

6 References 9

7 Conclusion 10

1
Acknowledgement

We would like to express our deepest appreciation to all those who provided us the
possibility to complete this report. A special gratitude we will give to our micro project guide,
Prof.T.R.Kawade who invested her full effort in guiding us in achieving the goal.

Furthermore we would also like to acknowledge with much appreciation the crucial role of the
class teacher Prof.T.R.Kawade who gave the permission to use all required equipment and the
necessary material to complete the micro project on “Case Study Of Fake Profiling” of subject
ETI

Our great obligation would remain towards our Head of Department Prof.B.B.Handge,
whose contribution in stimulating suggestions and encouragement helped us for writing report.
He provided with an opportunity to undertake the Micro project at Sir.Dr.M.S.Gosavi
polytechnic Institute Nashik. We appreciate the guidance given by other staff members of
Computer Engineering Department for Improving our presentation skills thanks to their
comment and advice.

We sincerely thanks to our respected Principal Dr.S.P.Deshpande proved to be a

constant motivation for the knowledge acquisition and moral support during our course
curriculum.

2
Abstract

In these project we are going to study the cases of fake profiling. In these firstly we are studying

 What is Digital Forensics?

 What are Fake Profiles?
 Who uses Fake Profiles?
 How we can detect Fake Profile?
 What we can do after detecting Fake Profile?
 How we can detect Fake Photo?

3
Case Study Of Fake Profiling
What are Fake Profiles?
A fake profile is the representation of a person, organization or company that does not truly exist,
on social media. Often these accounts use names and identities that not only look real but are
designed to get closer access to specific people and their target audience. The appearance of
these fake profiles can range from an attractive woman, who is trying to gain access to a man’s
Facebook, or a business such as a bank, reaching out to you for updated account information.
They usually are recently opened accounts that have few friends, anywhere from just a dozen to
several hundred. The pictures they use, are usually altered versions of images stolen from actual
people or organizations. So, who knows? maybe someone is using your pictures for malicious
purposes somewhere on the internet to deceive, gain access, and exploit your datam D

Who uses Fake Profiles?

Anyone can use a fake profile From government agencies, companies, kiddie hackers and
professional online criminals. All you need is a phone number, an email, a few stolen and altered
images, and you are good to go. Truly strong fake identities are built slowly and take time. No
one built their Facebook profile in one day, so that should be a clear red flag. Marketing agencies
also use them, as they want to research what people are doing online. It is not completely illegal
since users grant profiles access to their publicly posted information on social media, they are
granting them with full awareness and consent. Social engineering methods are used to surpass
any psychological barriers that the person behind the fake profile might detect in his targets. This
includes gender, age, ethnicity, appearance, where they work and went to school – all of these
components allow hackers closer access to their goal – your data.

4
How Can We Detect That a Profile is Fake?
If you think a fake profile is attacking you, it’s critical to look for signs of that profile being fake.
You could try reverse engineering profile images by searching for them on google images, but a
good hacker will know to alter them well enough the make them untraceable. So, what signs can
you expect to find?
1. The profile has very few pictures or no actual picture of a person at all.
2. It was created recently – in the past year or two. Unless the person is very young, most of
us opened our Facebook accounts in 2006-2007. Look for signs of long-term use of the
account.
3. Little or no contacts in common. When the profile has nothing in common with you such
as friends or even a professional interest, and they are trying to add you, it’s likely for
malicious reasons.
4. When a profile adds you but once you accept them, there is no more interaction with that
user. This is a classic sign that the profile is fake. The person behind the fake identity just
got what he/she wanted and will avoid talking to you in the hopes you forget about it.

What to do when fake profile is detected?

Report it. All social media websites now have the option of reporting a profile. Once you spot a
fake profile, be sure to report it, and remove it from your friend’s list. Whenever possible, be
safe and block the account to revoke its access to seeing your account. I recommend checking
your profile settings often and clean your contact list regularly from anyone you don’t know or
that you aren’t quite sure if they are real or not.

Not only can fake identities endanger your personal privacy, fake profiles can also put
companies in danger. For more information on how you can further improve your workforce to
avoid cyber threats, please visit our website at Cybint Solutions

positive rate and 1% false negative rate.

In [29], the authors used ground-truth
provided by RenRen
5
to train an SVM classifier in order to detect
fake accounts.
Using simple features, such as:
• frequency of friend requests
• fraction of accepted requests
The authors were able to train a classifier
with 99% true-
positive rate (TPR) and 0.7% false-positive
rate (FPR).
In [26], researchers used a ground truth
provided by Twitter;
the data have been processed using two
main approaches:
• Single classification rules
• Feature sets proposed in the literature for
detecting
spammers
Some features have been used from previous
work such as
Stateofsearch.com rule set [31], and
Socialbakers rule set
6
[32]. The authors were able to correctly
classify more than
95% of the accounts of the original training
se
The authors were able to classify the data
with 3% false
positive rate and 1% false negative rate.
In [29], the authors used ground-truth
provided by RenRen
to train an SVM classifier in order to detect
fake accounts.
Using simple features, such as:
• frequency of friend requests
• fraction of accepted requests
The authors were able to train a classifier
with 99% true-
positive rate (TPR) and 0.7% false-positive
rate (FPR).
In [26], researchers used a ground truth
provided by Twitter;

7
the data have been processed using two
main approaches:
• Single classification rules
• Feature sets proposed in the literature for
detecting
spammers
Some features have been used from previous
work such as
Stateofsearch.com rule set [31], and
Socialbakers rule set
[32]. The authors were able to correctly
classify more than
95% of the accounts of the original training
se
The authors were able to classify the data
with 3% false
positive rate and 1% false negative rate.
In [29], the authors used ground-truth
provided by RenRen
to train an SVM classifier in order to detect
fake accounts.
8
Using simple features, such as:
• frequency of friend requests
• fraction of accepted requests
The authors were able to train a classifier
with 99% true-
positive rate (TPR) and 0.7% false-positive
rate (FPR).
In [26], researchers used a ground truth
provided by Twitter;
the data have been processed using two
main approaches:
• Single classification rules
• Feature sets proposed in the literature for
detecting
spammers
Some features have been used from previous
work such as
Stateofsearch.com rule set [31], and
Socialbakers rule set
[32]. The authors were able to correctly
classify more than
9
95% of the accounts of the original traini
Feature Based detection
This approach relies on user-level activities
and its ac-
count details (user logs and profiles). Unique
features are
extracted from recent user activities (e.g.
frequency of friend
requests, fraction of accepted requests), then
those features
are applied to a classifier that has been
trained offline using
machine learning techniques [20], [27], [28],
[29].
In [27], the authors used a click-stream
dataset provided by
RenRen, a social network used in China
[30], to cluster user
accounts into similar behavioral groups,
corresponding to

10
real or fake accounts. Using the METIS
clustering algorithm
with both session and clicks features, such
as:
• Average clicks per session
• Average session length
• The percentage of clicks used to send
friend requests
• Visit photos
• Share c

Investigation
The investigation takes place through online mode where most of the potential evidence is stored
in OSN providers' databases. There is a variety of information that can be obtained through an
OSN account, and which is likely to be useful for an investigation. Such information includes the
user‟s profile containing full name, home address, telephone number, location, history of
education and work; links to other users with whom they have connections and relationships; and
also media being shared. The evidence and information searching process is an iterative process
whereby an investigator will search for the profiles of particular users, then profiles of these
users‟ friends, also profiles of friends of the users‟ friends and so on. The depth of iteration will
depend on the incident being investigated. To include those additional features, we have

11
proposed a new model that consists of four processes. The proposed model is shown in Figure 1.
The next section will discuss the activities in each of the processes in detail. A. Preliminary After
an incident occurs, the investigation will commence in the Preliminary Process. The purpose of
this process is to validate, measure, and plan the strategy that will be applied in the proceeding
processes. The regular method used by traditional digital forensic investigations involves
validation of the incident and assessment of the situation before developing an appropriate
strategy. Then the processis followed by checking on any requirements for the investigation such
as human resources, special equipment or legal resources. In OSN digital forensic investigations,
we will determine what to look for in the OSN site. Activities might involve planning which
profiles to initially focus on, determining the important connections between profiles, or
identifying what kind of information to look for in the profiles depending on the type of incident
being investigated. B. Investigation The Investigation Process consists of a number of activities
related to the OSN investigation. The aim of this process is to collect and store potential digital
evidence and the information required in order to proceed with the investigation. During this
process, the steps that will be involved will include the following. Online searching of users‟ or
targets‟ profiles, the link between targets that can contribute to useful information related to the
incident being investigated, and any potential evidence and information from the profile. Any
potential evidence found will be collected in a forensic manner to make sure that the evidence is
valid and presentable in a court of law or as part of any other relevant legal process or procedure.
These steps will be repeated if it becomes necessary to collect information and evidence from
more than one profile. Once the potential evidence has been found, it must be stored and the
evidence transported for analysis.

Analysis

The Analysis Process is a crucial part of the forensic investigation. In this process, the
investigator will need to verify that all evidence and information found is connected to the
incident being investigated. The traditional activity during the analysis process usually involves
the examination and analysis of evidence to determine its value and impact. In OSN digital
forensic investigations

12
References:-

www.google.com

http://www.cplusplus.com/

13
Conclusion:-

Hence we have studied the case of fake profiling

14
15