Working title

Is it easy for fake accounts to gain the information they need to impersonate/fraud another
user?

Intro
When looking to the how fake users on social media are common it was easy to see how the
can impersonate a person. Also when looking at whether if it is illegal to impersonate a person it
only became clear it was when the user was being fraud. The main issue with fake users is that
they are hard to detect and also difficult to be found. There is limited research that can be found
on the topic. Which in turn means there is a lot that can be explored when looking at what can
be done to help stop the use of fake accounts. There are possible applications for the sites that
could be suggested such as a program which will detect if the information and images have
been used before of the social media site that the fake account would like to use.

Research Question, Aims, Objectives and Deliverable

The overall aim for the project is identify the main problem with impersonation online via social
media applications. Whether this be by how the fake account users the information that they can
gather publicly or if it is the information that they can gather to fraud a person. There are
multiple objectives that will need to be carried out to gain the answers for the project. These
include conducting a survey that would allow social media users to give a perception on whether
they know if social media sites are aware of what fake accounts are. And also looking at the
data from social media applications to see if there is any duplicates in the data that they have.
To solve the issue at hand an application would need to be developed to stop fake accounts
with user’s information been given access to the social media site.

Literature Review

Detecting Clusters of Fake Accounts in Online Social Networks

In addition to the ever growing use of social media with billions of user across all social media
sites, there will be the users that are legitimate users and those that are not. Those are called
Fake accounts and can be used for many different reasons. One of the main reasons fake
accounts are made is for fraud. This can be done in many ways. But it is important that
detecting and taking action is fast, to protect the real users from having their account hacked.
Although for a social media user it might be hard to detect a really user as they can appear to
be so real. The researchers for this paper have used an offline machine to detect patterns within
the information a user will enters to detect if they are real or not. This could be the email
address that they use or the name that the enter. The framework used was to create Cluster
Builders which takes a raw list of accounts along with their raw features. It look at all the
accounts which have the same IP Address and this is one of the ways they are able to find
which accounts are fake. Profile Featuriser is a key component the whole purpose of it is to
convert the raw data for each cluster and place them into a single numerical vector. The results
from the paper show that on LinkedIn were approximately 275,000 users and over half of those
were labelled as been fake or spam accounts. This paper I found is a good example of how fake
accounts can appear and how easy it is for them to appear real. It also a good paper for the
topic I have chosen as it gives a clear idea of what needs to be done to protect against fake
accounts.

Social Snapshots: Digital Forensics for Online Social Networks

This research paper looks at how the law enforcement and academia collect data from social
media sites. A social snapshot is used to collected the datasets which include the users data
and related information which all comes from social media sites. This paper also looks at using
the social snapshot to do the data harvesting that is required. This paper discusses a hybrid
system that uses and automated web browser. Also combining the social snapshot with the third
party hybrid system is initialised with the users credentials or cookies. The framework for this
paper looks at using the social snapshot technique which targets certain users via credentials or
cookies, this is done via an authentication mechanism. From this is add an application to the
targeted users and send an API to a secret application server. The data is then stored. The data
that is collected via the third party application is stored on the Social Snapshot application
server. With this application over 200 HTTP requests were Snapshot in less than 15 minutes.
This paper relates to the topic I have chosen as it looks at how easy it is to store and gather
information in a short amount of time which can be seen in the results that were collected. It is
also interesting how a snapshot can show a lot of information and that a hybrid of two
technologies can be used to gather information from social media rather than just using one
simple piece of technology.

Millions lost in 'celebrity-backed scams'

Many celebrities are endorsing a new currency which in turn has become a part of a fraud. The
influencers which are promoting this new currency have a large social following this leads to a
world-wide search for potential victims to fraud. From FCA (Financial Conduct Authority) it is
said that endorsements are made through expensive items such as watches and cars on social
media which is used to promote the scheme that the person who wants to fraud will use to
advertise. In order for the scheme to appear more legit the frauders lead to what appears to be
a real website. Where people that have arrived there are pressured into an investment. The
crypto currency which is been investigated is bitcoin. This sort of currency is only available
online and has no central banking system. Laura Suter, from AJ Bell, said: “If your friend
recommends something to you, don't assume that they have done all the research for you.”
This comes as the crypto currency had tripled with fraud in 2018 with close to 1834 cases with
each victim losing more the £14,000. This comes as a shock as people are not aware of where
their money is going to. They should always do research to find out where the money that they
are investing into is going to where it claims to be. There has been many warning to stop the
“get rich quick” schemes with little success. Thoses that are investing into the currency have
very little understand of where they are actually sending the money and what it is been used for.
Due to the internet currency there is no protection or compensation on how to protect the
investor from this happening again. I feel that this article from the BBC relates to the topic I have
chosen as it looks how people on social media are able to use the influence they have to
pressure their followers into buying into a scheme that they know nothing about. It is important
that members of the public know how easy to lose a lot of money within a matter from an advert
from social media. It also shows not only can people grab your personal details from social
media but that they can also pressure you into doing something that could harm the user in
other ways.

Research design
In order for a question a design needs to be put in place so that you do not lose track of where
you are going. So this is how a project will be investigated. For the dissertation the aim is to look
at what research needs to be done.

For the dissertation to take a full effect data first needs to be collected so that there is some
form of start to this project. Without collecting data it is hard part of the research project. This
data needs to be analysed as clean so that any data that is not needed or nulls are not there.
This may take a while to do as a lot of data may be collected and therefore make take a lot of
time to complete. Through a literature review it will come clear what methodologies that have
been used have worked to some degree. A literature review will be complete to look at other
research papers and how they have dealt with an issue similar to that of the topic of fraud and
social media. It is important that the literature review has points that relate to the topic and that it
can be linked back to previous research that has been completed. There is little knowledge on
fake accounts on social media and also on that of fraud. It is necessary that the dissertation that
is to be completed shows a methodology that has not been done before that would ideally stop
both fake accounts and fraud on social media sites. To gather the data needed a questionaire
will be created and distributed via different sources this will be done face to face and over social
media sites.

The choice to use qualitative data was made as the data that will be collected would either be
through interviews and questionnaires rather than a database of information. The only issue
with conducting interviews and questionnaires is that members that compete theses may not be
completely honest and not give the answer honestly. With the topic chosen been specific
members of the public may not know much information on what fraud is and fake accounts. This
could lead to them been misinformed. This then could lead to the data been collected not being
correct. To stop this from happening before the interview or questionnaire is conducted a brief
summary would be given to stop the any confusion that might occur. This would give be a
definition of what fraud is and in what ways it can occur. And similar with fake accounts on
social media. It is important that each participant knows what the definition and background of
the issues is. The questionnaires and interviews are the main where data will be collected from,
as well as looking at databases that my be able to help. This type of data would be used to
more for the fake accounts from social media as it would be easier for members of the public to
be able to answer questions about social media rather than that of fraud.

When looking at collecting data that has already been collect it can be difficult to clean the data
to find if it is useful for the topic. The data that would be collected from a database would be
quantitative data. As this would be reported as part of a statistically report with the numbers with
comparisons been made throughout the report. Unlike the qualitative data this dataset would
already be clean and able to use right away. No issues should arise while using this dataset,
also no human interaction would be needed for this data as it had already been collected
meaning that the data would be correct and with no issues. When collecting this type of data it
would be for the fraud aspect of the dissertation as a more wide range of data would be
available to use. With fraud it is a topic that is a topic which is spoke about, but not many people
actual know what it is.
The aim of the dissertation is to give a conclusive analysis, looking at all of the elements that
occur with the topic which has been chosen. This will include a suitable methodology which will
help to give direction and what will help to aid question at hand. It is important that all aspects of
the dissertation are discussed to a certain degree and are able to provide an answer which
would help to answer the research question.

Ethics
There are multiple resources that could be involved in order to complete the research that is at
hand. For example, a questionnaire would be filled out for the gaining data about whether
members of the public are aware about fake accounts on social media and this could lead to
misunderstanding data. Another risk to the survey would be if it is conducted outside in an
uncontrolled environment. This could lead to the survey conductor coming into harm’s way and
injuring themselves. This could be meaning that there is not a good representation of the data
that is needed. There are also issues that may arise when wanting to gather data from social
media sites as it may be difficult to gain the access that is needed. This could lead to legal
issues been raised and may not be given access to the data. Another issue is that gaining data
that is publicly available may also have missing data and this is an issue because it can give out
missing results which may affect the whole project.

Time Plan and Gantt Chart

The gantt charts starts in September of 2019, due to having a placement that starts as of July
2019. There is around 15 weeks in which I have stated the dissertation would be completed in.
this starts at the literature review right up to the the concluding of the overall dissertation. But
due to the placement it means I have a little more time to do research for the paper.
Reference
Xiao, C., Freeman, D. and Hwa, T. (2019). Detecting Clusters of Fake Accounts in Online Social
Networks.

Anon, (2019). [online] Available at:

http://scholar.google.com/citations?user=MzMmsi8AAAAJ&hl=en

BBC News. (2019). Millions lost in 'celebrity-backed scams'. [online] Available at:
https://www.bbc.co.uk/news/business-
48336169?intlink_from_url=https://www.bbc.co.uk/news/topics/cvwydw4g8pzt/fraud&link_locati
on=live-reporting-story

Ethics Checklist

RESEARCH ETHICS CHECKLIST FOR STUDENTS (SHUREC7)

This form is designed to help students and their supervisors to complete an ethical
scrutiny of proposed research. The SHU Research Ethics Policy should be consulted
before completing the form.

Answering the questions below will help you decide whether your proposed research
requires ethical review by a Designated Research Ethics Working Group.

The final responsibility for ensuring that ethical research practices are followed rests with
the supervisor for student research.

Note that students and staff are responsible for making suitable arrangements for keeping
data secure and, if relevant, for keeping the identity of participants anonymous.
They are also responsible for following SHU guidelines about data encryption and
research data management.

The form also enables the University and Faculty to keep a record confirming that
research conducted has been subjected to ethical scrutiny.

For student projects, the form may be completed by the student and the supervisor and/or
module leader (as applicable). In all cases, it should be counter-signed by the supervisor
and/or module leader, and kept as a record showing that ethical scrutiny has
occurred. Students should retain a copy for inclusion in their research projects, and staff
should keep a copy in the student file.

Please note if it may be necessary to conduct a health and safety risk assessment for the
proposed research. Further information can be obtained from the Faculty Safety Co-
ordinator.

General
Details

Name of student Student

SHU email address

Course or qualification
(student)

Name of supervisor

email address

Title of proposed research Is it easy for fake accounts to gain the information
they need to impersonate/fraud another user?

Proposed start date

 Proposed end date
Brief outline of research to
include, rationale & aims
(250-500 words).
Where data is collected
from individuals, outline
the nature of data, details
of anonymisation, storage
and disposal procedures if
required (250-500 words).
1. Health Related Research Involving the NHS or Social Care / Community Care or the
Criminal Justice Service or with research participants unable to provide informed
consent
Question
When looking to the how fake users on social
media are common it was easy to see how the can
impersonate a person. Also when looking at
whether if it is illegal to impersonate a person it
only became clear it was when the user was being
fraud. The main issue with fake users is that they
are hard to detect and also difficult to be found.
There is limited research that can be found on the
topic. Which in turn means there is a lot that can
be explored when looking at what can be done to
help stop the use of fake accounts. There are
possible applications for the sites that could be
suggested such as a program which will detect if
the information and images have been used
before of the social media site that the fake
account would like to use.
Yes/No
1. Does the research involve? no

• Patients recruited because of their past or present use of

the NHS or Social Care
• Relatives/carers of patients recruited because of their
past or present use of the NHS or Social Care
• Access to data, organs or other bodily material of past or
present NHS
patients
• Foetal material and IVF involving NHS patients
• The recently dead in NHS premises
• Prisoners or others within the criminal justice system
recruited for health- related research*
• Police, court officials, prisoners or others within the
criminal justice system*
• Participants who are unable to provide informed
consent due to their incapacity even if the project is
not health related

2. Is this a research project as opposed to service evaluation

or audit?
For NHS definitions please see the following website
http://www.hra.nhs.uk/documents/2013/09/defining-research.pdf

If you have answered YES to questions 1 & 2 then you must seek the appropriate
external approvals from the NHS, Social Care or the National Offender
Management Service (NOMS) under their independent Research Governance
schemes. Further information is provided below.

NHS https://www.myresearchproject.org.uk/Signin.aspx

* All prison and probation projects also need HM Prison and Probation Service (HMPPS)
approval. Further guidance at:
https://www.myresearchproject.org.uk/help/hlphmpps.aspx

NB FRECs provide Independent Scientific Review for NHS or SC research and initial
scrutiny for ethics applications as required for university sponsorship of the research.
Applicants can use the NHS proforma and submit this initially to their FREC.

2. Research with Human

Participants

Question Yes/No
Does the research involve human participants? This includes yes
surveys, questionnaires, observing behaviour etc.

Question Yes/No

1. Note If YES, then please answer questions 2 to 10

If NO, please go to Section 3

2. Will any of the participants be vulnerable?

Note: Vulnerable’ people include children and young people, people
with learning disabilities, people who may be limited by age or
sickness, etc. See definition on website

3. Are drugs, placebos or other substances (e.g. food substances,

vitamins) to be administered to the study participants or will the study
involve invasive,
intrusive or potentially harmful procedures of any kind?

4. Will tissue samples (including blood) be obtained from no

participants?

5. Is pain or more than mild discomfort likely to result from the study? no

6. Will the study involve prolonged or repetitive testing? no

7. Is there any reasonable and foreseeable risk of physical or emotional no

harm to any of the participants?
Note: Harm may be caused by distressing or intrusive interview
questions, uncomfortable procedures involving the participant, invasion
of privacy, topics relating to highly personal information, topics relating
to illegal activity, etc.

8. Will anyone be taking part without giving their informed consent? no

9. Is it covert research?
Note: ‘Covert research’ refers to research that is conducted without the
knowledge of participants.
10. Will the research output allow identification of any individual no
who has not given their express consent to be identified?

If you answered YES only to question 1, the checklist should be saved and any course
procedures for submission followed. If you have answered YES to any of the other
questions you are required to submit a SHUREC8A (or 8B) to the FREC. If you answered
YES to question 8 and participants cannot provide informed consent due to their
incapacity you must obtain the appropriate approvals from the NHS research governance
system. Your supervisor will advise.

3. Research in
Organisations

Question Yes/No

1. Will the research involve working with/within an organisation (e.g. no

school, business, charity, museum, government department,
international agency, etc.)?

2. If you answered YES to question 1, do you have granted access

to conduct the research?
If YES, students please show evidence to your supervisor. PI should
retain safely.

3. If you answered NO to question 2, is it

because:
A. you have not yet asked
B. you have asked and not yet received an answer
C. you have asked and been refused access.
Note: You will only be able to start the research when you have been
granted access.

4. Research with Products and Artefacts

Question Yes/No
1. Will the research involve working with copyrighted documents, yes
films, broadcasts, photographs, artworks, designs, products,
programmes, databases, networks, processes, existing datasets
or secure data?

2. If you answered YES to question 1, are the materials you intend to use yes
in the public domain?
Notes: ‘In the public domain’ does not mean the same thing as ‘publicly
accessible’.
• Information which is 'in the public domain' is no longer
protected by copyright (i.e. copyright has either expired or been
waived) and can be used without permission.
• Information which is 'publicly accessible' (e.g. TV broadcasts,
websites, artworks, newspapers) is available for anyone to
consult/view. It is still protected by copyright even if there is no
copyright notice. In UK law, copyright protection is automatic
and does not require a copyright statement, although it is
always good practice to provide one. It is necessary to check
the terms and conditions of use to find out exactly how the
material may be reused etc.
If you answered YES to question 1, be aware that you may need to
consider other ethics codes. For example, when conducting Internet
research, consult the code of the Association of Internet Researchers;
for educational research, consult the Code of Ethics of the British
Educational Research Association.

3. If you answered NO to question 2, do you have explicit permission

to use these materials as data?
If YES, please show evidence to your supervisor.

4. If you answered NO to question 3, is it A/B/C

because:
A. you have not yet asked permission
B. you have asked and not yet received and answer
C. you have asked and been refused access.
Note You will only be able to start the research when you
have been granted permission to use the specified material.

Adherence to SHU policy and procedures

 Personal statement

I can confirm that:

− I have read the Sheffield Hallam University Research Ethics Policy and Procedures
− I agree to abide by its principles.

Student

Name: Date:

Signature:

Supervisor or other person giving ethical sign-off

I can confirm that completion of this form has not identified the need for ethical
approval by the FREC or an NHS, Social Care or other external REC. The research
will not commence until any approvals required under Sections 3 & 4 have been
received.

Name: Date:

Signature:

Additional Signature if required:

Name: Date:

Signature:

Please ensure the following are included with this form if applicable, tick box to indicate:
 Yes No N/A

Research proposal if prepared previously

Any recruitment materials (e.g. posters,

letters, etc.)

Participant information sheet

Participant consent form

Details of measures to be used (e.g.

questionnaires, etc.)

Outline interview schedule / focus group

schedule

Debriefing materials

Health and Safety Project Safety Plan for

Procedures