Title: A hybrid approach for social media forensics

Authors: Amir Aboubakr Shaker Mahmoud; Ngaira Mandela; Nilay R. Mistry; Animesh Kumar Agrawal

Addresses: School of Doctoral Studies and Research (Digital Forensics and Cyber Security), National Forensic Sciences University, Gandhinagar, India ' School of Doctoral Studies and Research (Digital Forensics and Cyber Security), National Forensic Sciences University, Gandhinagar, India ' School of Doctoral Studies and Research (Digital Forensics and Cyber Security), National Forensic Sciences University, Gandhinagar, India ' Unitedworld Institute of Technology (UIT), Karnavati University, Ahmedabad, India

Abstract: Social media plays a pivotal role as a rich source of data for digital forensics investigators, enabling the extraction of valuable evidence for crime analysis. This research introduces a hybrid digital forensics approach for social media investigation, compatible with all web-browsing devices. This hybrid approach combines a three-stage classification process consisting of TextBlob sentiment analysis to analyse the tweet's polarity, VADER to detect suspicious tweets, and a crime-type dictionary to categorise crime-related tweets. Using Python libraries, Twitter data is collected using authentic Twitter accounts, hashtags, or keywords, then cleaned, translated, geolocated, and classified. A real-time interactive platform is implemented for crime detection and analysis, bolstering the capabilities of law enforcement agencies and researchers in understanding crime patterns. The research concludes with promising results, highlighting the potential of the approach, and discusses future enhancements, ultimately aiding in crime analysis and prevention.

Keywords: crimes analysis; cybercrimes; digital forensics; hybrid methods; text blob; VADER; social network analysis; Twitter; law enforcement.

DOI: 10.1504/IJESDF.2025.147160

International Journal of Electronic Security and Digital Forensics, 2025 Vol.17 No.4, pp.460 - 483

Received: 10 May 2023
Accepted: 21 Sep 2023
Published online: 11 Jul 2025 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article