Multi-Dimensional Data Analysis for Offender Classifications on Social Media using Big Data Techniques

Main Article Content

Wongyos Keardsri


The objectives of this research are 1) to study the characteristics of data in the dimensions of personal information, personal behaviors, and commentary information of offenders on social media; 2) to investigate, track, record, and analyze the data of offenders on social media by using statistical Big Data techniques and 3) to classify social media offenders by using statistical Big Data techniques. The results of the study found that there were 4,974 Twitter accounts of social media offenders and could be used to count the keywords from every account's hashtag in 1,035 keywords. The results of the python program are able to classify offenders on social media showed that there were 11 of the Twitter groups of social media offenders, while the highest number of social media accounts of offenders committing illegal activities related to pirated products was at 19.44%, sex toys at 16.32%, prostitution at 13.73%, illegal firearms at 9.95%, online gambling at 9.17%, informal loans at 7.64%, illegal drug at 6.94%, forged documents at 5.19%, narcotics at 4.70%, other crimes at 4.66, and wildlife trading at 2.25%, respectively. Moreover, the results of the correlation of the offender information on social media by using an open-source intelligence tool with the Maltego could be used to effectively search social media accounts of perpetrators in the police work.


Download data is not yet available.

Article Details

How to Cite
Keardsri, W. (2021). Multi-Dimensional Data Analysis for Offender Classifications on Social Media using Big Data Techniques. Journal of Criminology and Forensic Science, 7(1), 191-203. Retrieved from
Research Articles


Chena, R and Sharmaa, S. K. (2013). Understanding Member Use of Social Networking Sites from a Risk Perspective. Procedia Technology, 9, 331-339.

Cooper, A. (2016). Using social media for social research: An introduction. Social Media Research Group, Retrieved September 2, 2020. From

Hassan, N. (2018). A primer introduction to Open-Source Intelligence Gathering (OSINT) gathering and analysis. Retrieved September 2, 2020. From

Hayes, D. R. and Cappa, F. (2018). Open-source intelligence for risk assessment. Business Horizons, 61(5), 689-697.

Khemakapasiddhi, C., Nurarak, P., Pumsorn, P., Ramungtong, P., Thongyod, E., and Keardsri, W. (2019). Personal Relationship Analysis of Prostitution and Human Trafficking Data on Twitter using Open-Source Intelligence Tools. Senior Research Project, Bachelor of Public Administration Program in Police Science, Royal Police Cadet Academy. (In Thai).

Riek, M. and Böhme, R. (2014). Understanding the influence of cybercrime risk on the e-service adoption of European Internet users. Proceedings of 13th Annual Workshop on the Economic of Information Security.

Sathyadevan, S., Devan M. S., and Gangadharan, S. (2014). Crime Analysis and Prediction Using Data Mining. Proceedings of First International Conference on Networks & Soft Computing (ICNSC).

Smitherson, D. (2012). Impact of Cyber Crime and Security on Social Media. Social Media Today, Retrieved September 2, 2020. From

Socialbakers. (2020). Free Social Media Statistics. Retrieved September 2, 2020. From https://www.socialbakers .com/statistics/.

UNHCR. (2017). Social Media and Forced Displacement: Big Data Analytics & Machine-Learning. Retrieved September 2, 2020. from

Williams, M. L., Burnap, P. and Sloan, L. (2017). Crime Sensing with Big Data: The Affordances and Limitations of using Open Source Communications to Estimate Crime Patterns. British Journal of Criminology, 57 (2), 320-340.