Professional Documents
Culture Documents
Analysis of Cyber Crime On Social Media
Analysis of Cyber Crime On Social Media
net/publication/351945517
CITATIONS READS
14 7,855
4 authors:
All content following this page was uploaded by Said A. Salloum on 28 May 2021.
1 Introduction
The sphere of cyberspace has reached everywhere on the globe [1]. The advancements
in cybersecurity in the fields of technologies, activities, and methods for protecting the
various programs and networks and data associated with computers thereby prevent
attacks by malicious people and/or cybercriminals. The eye-opener is that cybersecurity
does not entirely prevent cyberattacks. It is stated that millions of these cyberattacks
could be proliferated onto cyberspace whose analysis would turn up complicated [2].
This posts severe necessity to stay alert on such attackers’ prevalence and protect oneself
or network against such activities where such an awareness alone does not hold any good
an implementation point of view. Online social networks gained popularity tremendously
in recent years. The social media platforms could be explained as “a cluster of Internet-
based applications that team up on the foundations of Web 2.0 which allows creating
and exchanging user-generated content” [3]. Social media platforms have become an
integral part of people’s lives, including but not limited to LinkedIn, YouTube, Facebook,
Snapchat, Instagram, etc., [4]–[11]. These are highly dependent as they aid in building
social relationships and building strong communities online, which is considered an asset
for further growth of individuals or groups [12]. These web-based platforms render varied
novel opportunities for socializing and interacting with the users that have redefined
the approach of sharing information such as online public based opinions on polls,
news circulation, the successful conduct of online businesses, and/or promotion to a
larger crowd of the audience than the traditional methodologies. People are becoming
obvious targets of cybercriminals through social networks due to such platforms’ system
design. To safeguard social networks, various organizations are constantly increasing
the Information Technology (IT) budgets for protecting these social networks around
the globe [13]. The voluminous amount of data being shared through social media are
collected and analyzed for finding different patterns to get useful insights into the data
[14]. Thus, it is highly essential to follow modern techniques and tools for sealing the
resources and information available online via social networks.
2 Related Work
An analysis of various cyber crimes on Facebook has been performed by [15, 16] by
comparing the website’s penetration in 148 different countries. The analysis has used
Ordinary Least Squares (OLS), Tobit, and Quantile-based regressions. The data used
contains details of the population from different regions and their income levels to
identify the link between the crimes and the financial status. The analysis shows that there
is not much correlation between the income status and the crime rates, especially in Arab
regions. However, the correlation was positive in sub-Saharan Africa. The similarities
between conventional crimes and cybercrimes have been compared by [17] by analyzing
the crime rates and their reporting behavior. The data was gathered through a survey
conducted by Dutch research. It is seen that conventional crimes are widely reported in
comparison with cybercrimes. Only high-level cybercrimes are usually reported, while
others are usually ignored. This shows that there are a lot of cybercrimes taking place
which never get recorded and reported.
The link between the different types of crimes has been analyzed by [18]. The analysis
was conducted using Machine learning algorithms on the large data that includes the
data collected from Twitter and police survey data in order to identify the link between
the online hate speeches concerning religion and race and physical crimes taking place.
Similar analysis has also been performed by [19], where the data has been collected from
Twitter messages and analyzed with the region’s physical violence. While the former
Analysis of Cybercrime on Social Media Platforms 617
research analyzed racial and religious messages, the latter analyzed the violence taking
place during sporting events between hockey and basketball fans. Historical data has
also been used for understanding the previous instances of crimes. The research also
shows a high correlation between tweets and violence. The police can monitor these
tweets to predict if any violence will occur in their city. Both these researches have used
spatial data with geotagged tweets to identify the local tweets.
3 Methodology
3.1 Research Questions
This research aims to understand the characteristics of cybercrimes in social media and
identify what kind of efforts the police should take to control cybercrime. The research
strives to answer the following research questions:
• What types of attacks, vulnerabilities, and crimes are taking place in social media?
• What are the demographics of most of the criminals?
• What kind of efforts should the police take to monitor the criminals in social media
and reduce crime?
• This 12 GB RAM.
• Core i3 processor.
• Python 3.7 Jupyter notebook console.
Python gives various graphing libraries which are packed with a wide range of features
according to user preferences [20]. The library’s excellent support by Python makes it
interesting to create highly specialized and supported plots in Python. The overview of
each plotting libraries are given below for detailed understanding:
3.3.1 Seaborn
Seaborn provides a wide range of attractive graphs created on Matplotlib that is consid-
ered a Python data visualization library. Seaborn is used with great flexibilities. Graphs
are created in a single line which is equivalent to ten lines in Matplotlib. The design stan-
dards are highly impressive that go well while working with the panda’s data frames.
Seaborn could be imported by typing the following:
import seaborn as sns.
3.3.2 Matplotlib
The most popular and widely used Python plotting library is the Matplotlib, a low-level
library that could be interfaced with Matlab that renders high degrees of freedom when
it comes to creating larger code. The Matplotlib installation through pip and Conda are
as follows:
Pip install matplotlib or condainstall matplotlib.
Matplotlib is designed for the creation of all the basic graphs such as line chart, bar
chart, histogram, etc., and it could be imported as follows:
Import matplotlib.pyplot as plt.
Both the above-stated libraries have been used for the visualization of the dataset.
3.4 Data
3.4.1 Dataset
The analysis of cybercrime on the social media platforms in the form of offensive threats
was carried out with the help of the social media cyber crime dataset, which consisted of
574 observations [21] (See Appendix A). Each of those observations was a measure for
the suspected person arrested for committing any form of cybercrime on social media
platforms incorporating the background details of these features from the year 2014 to
2018. The following are the background features pertaining to the suspects under control
which are given in Table 1.
4 Results
This section consists of graphs and results, which are performed in the methodology
section by using the exploratory data analysis method. Figure 1 shows the crimes in
social media that have been taken year-wise from 2014 to 2018. Most of the crimes took
place in 2018 with 28.1% of the crimes, followed by 20.1% in 2015. The proportion of
crimes in the other years is 18.3%, 17.5%, and 16.1% in 2017, 2016, and 2014.
Figure 2 shows the different types of crimes taking place in social media. From
the data, 300 instances of fraud were present, followed by around 100 instances of
620 A. Almansoori et al.
child pornography. Other crimes are significantly lower with around 50 cases each for
threatening, criminal harassment, and child lurking.
The plot in Fig. 3 represents the previous criminal records of the criminals. It is seen
that most of the criminals have a previous record of criminal history. Out of the total
data, around 70% of them had a previous criminal background, while 30% did not have
any criminal background.
Analysis of Cybercrime on Social Media Platforms 621
The educational background of the suspects is shown in Fig. 4. Most of the criminals
did not have high education. Around 70% of the criminals did not have any type of
education or only had basic education. 30% of them were graduates.
Figure 5 depicts the financial background of criminals. Around 61% of the criminals
were very poor, while the rich and the middle case made up the other 39% population.
622 A. Almansoori et al.
The data in Fig. 6 is a representation of the distribution of the age of the suspects.
Most of the criminals were between the ages of 20 and 25 with a peak at 22 years. There
are hardly any criminals above the age of 45.
The analysis of criminal background with educational background throws light upon
the various grey areas in the society that needs immediate addressing to create harmony
in the surroundings, as shown in Fig. 7.
Analysis of Cybercrime on Social Media Platforms 623
5 Conclusion
Cybercrime must be the primary focus of law enforcement agencies that should provide
sufficient resources for the same. The usage of social media platforms for interacting
with peers and others by billions of users worldwide has captivated the cybercriminals
to spread malicious code, inbox users with spam messages, trying to overcome users’
inherent trust in these platforms on the relationship and behavioral network. Thus, social
media networking platforms’ privacy and security models are recently becoming a major
concern for these users. The online social media sites need to identify the core aspects
of human and social connectivity through the interaction and proper identification of
accurate intuitive and then strong methodologies for ensuring all those essential lev-
els of maintaining privacy, granting protection, and building the trust framework such
tiers. The various governments, along with intelligence services teamed up with highly
skilled and technically strong experts, must be formed to adopt and frame various new
technologies along with the paradigms without disturbing the load should be made. The
amount of data flowing on such networks shall then be analyzed in this manner. In addi-
tion to this, cybersecurity should not be scared about; instead of proper planning and
implementing strategies to tackle the problem should be focused on. The integral part
should be assembled with proper analysis mechanisms from the users at all levels with
proper guidelines from various stakeholders is mandatory. As an individual, the safe
practices on the social media platforms could be made clear to various people who are
using them. Awareness from individuals to individuals will have a greater impact on the
problem. Thus, cybercrime analysis on social media platforms gave a clear picture that
individuals are the strong ties who are uplifting the bigger outlines. Hence, maintaining
security and privacy on such platforms is essential. It is seen that uneducated and poor
people have more tendency to commit cybercrimes. Also, the ages between 20 and 25
commit more crimes. Hence, this would help the police to identify the type of people
who would create more crimes.
624 A. Almansoori et al.
The police must focus on those with these factors while monitoring social media.
While this research has strived to identify the social media cybercrime and the police
efforts to mitigate these crimes, the samples considered are less due to time limitations.
In the future, this project can be expanded into a fully-fledged analysis by using Artificial
Intelligence techniques for the analysis. Also, more data can be considered by merging
data from different police departments, and the analysis can be conducted on a global
scale. The number of factors is limited in this work. Numerous factors affect the criminal
tendency to commit crimes. These factors can be identified in the future, and data may be
collected accordingly. Identification of such factors can help the police to narrow down
the criminals further.
References
1. Salloum, S.A., Alshurideh, M., Elnagar, A., Shaalan, K.: Machine learning and deep learning
techniques for cybersecurity: a review. In: Joint European-US Workshop on Applications of
Invariance in Computer Vision, pp. 50–57 (2020)
2. Jang-Jaccard, J., Nepal, S.: A survey of emerging threats in cybersecurity. J. Comput. Syst.
Sci. 80(5), 973–993 (2014)
3. Kaplan, A.M.: Social media, the digital revolution, and the business of media. Int. J. Media
Manag. 17(4), 197–199 (2015)
4. Chy, M.K.A., Ahmed, S.A., Doha, A.H., Masum, A.K.M., Khan, S.I.: Social media user’s
safety level detection through classification via clustering approach. In: 2019 International
Conference on Computer, Communication, Chemical, Materials and Electronic Engineering
(IC4ME2), pp. 1–4 (2019)
5. Salloum, S.A., Al-Emran, M., Monem, A.A., Shaalan, K.: A survey of text mining in social
media: Facebook and Twitter perspectives. Adv. Sci. Technol. Eng. Syst. J 2(1), 127–133
(2017)
6. Salloum, S.A., Al-Emran, M., Shaalan, K.: Mining social media text: extracting knowledge
from Facebook. Int. J. Comput. Digit. Syst. 6(2), 73–81 (2017)
7. Habes, M., Alghizzawi, M., Khalaf, R., Salloum, S.A., Ghani, M.A.: The relationship between
social media and academic performance: Facebook perspective. Int. J. Inf. Technol. Lang.
Stud. 2(1) (2018)
8. Salloum, S.A., Al-Emran, M., Habes, M., Alghizzawi, M., Ghani, M.A., Shaalan, K.:
Understanding the impact of social media practices on E-learning systems acceptance. In:
International Conference on Advanced Intelligent Systems and Informatics, pp. 360–369
(2019)
9. Alghizzawi, M., Habes, M., Salloum, S.A., Ghani, M.A., Mhamdi, C., Shaalan, K.: The effect
of social media usage on students’e-learning acceptance in higher education: a case study from
the United Arab Emirates. Int. J. Inf. Technol. Lang. Stud. 3(3) (2019)
10. Alghizzawi, M., Salloum, S.A., Habes, M.: The role of social media in tourism marketing in
Jordan. Int. J. Inf. Technol. Lang. Stud. 2(3) (2018)
11. Salloum, S.A., Mhamdi, C., Al Kurdi, B., Shaalan, K.: Factors affecting the adoption and
meaningful use of social media: a structural equation modeling approach. Int. J. Inf. Technol.
Lang. Stud. 2(3), 96–109 (2018)
Analysis of Cybercrime on Social Media Platforms 625
12. Icha, O., Agwu, P.E.: Effectiveness of social media networks as a strategic tool for
organizational marketing management. J. Internet Bank Commer, S2 (2015)
13. Ye, X., Zhao, B., Nguyen, T.H., Wang, S.: Social media and social awareness. In: Manual of
Digital Earth, Singapore, pp. 425–440 (2020)
14. Sharma, B.K., Joseph, M.A., Jacob, B., Miranda, L.C.B.: Emerging trends in digital forensic
and cyber security-an overview. In: Sixth HCT Information Technology Trends (ITT) 2019,
pp. 309–313 (2019)
15. Asongu, S., Nwachukwu, J., Orim, S.M., Pyke, C.: Crime and social media. Inf. Technol.
People 32(5), 1215–1233 (2019)
16. Al Mansoori, S., Almansoori, A., Alshamsi, M., Salloum, S.A., Shaalan, K.: Suspicious
activity detection of Twitter and Facebook using sentimental analysis
17. van de Weijer, S.G.A., Leukfeldt, R., Bernasco, W.: Determinants of reporting cybercrime:
a comparison between identity theft, consumer fraud, and hacking. Eur. J. Criminol. 16(4),
486–508 (2019)
18. Williams, M.L., Burnap, P., Javed, A., Liu, H., Ozalp, S.: Hate in the machine: anti-black
and anti-muslim social media posts as predictors of offline racially and religiously aggravated
crime. Br. J. Criminol. 60(1), 93–117 (2020)
19. Ristea, A., Al Boni, M., Resch, B., Gerber, M.S., Leitner, M.: Spatial crime distribution and
prediction for sporting events using social media. Int. J. Geogr. Inf. Sci. 1–32 (2020)
20. The python graph gallery (2017)
21. S. Canada, “Statistics Canada” (2018)