Identifying Covert Criminal Networks from Noisy and Sparse Text Data – UROP Spring Symposium 2025

Identifying Covert Criminal Networks from Noisy and Sparse Text Data

Vishalakshi Meyyappan

Research Mentor(s): Brinda Gokul
Mentor Department: Information
Authors: Brinda Gokul, Vishalakshi Meyyappan
Session: Session 3 (11:00am – 11:50am)
Presentation Type: Poster 106

Abstract

This research explores how criminal groups recruit individuals by analyzing 240 criminal records from five countries. These groups often use hidden methods, making it difficult to understand how they recruit new members. The study aims to uncover these hidden strategies by analyzing noisy and sparse text data and using methods like natural language processing and social network analysis to reveal patterns and relationships. A key question this research addresses is why lone offenders, who work independently, are seen as a significant concern by the FBI. While these offenders may act alone, they can still be connected to larger criminal networks in indirect ways. This study looks at how these individuals might be influenced through decentralized communication, digital platforms, and the spread of ideas. Previous research has mostly focused on more obvious recruitment strategies or organizational structures but has not fully explored the hidden mechanisms behind recruitment. This research fills that gap by studying the covert connections and applying new methods to gain a better understanding of how criminal groups operate. The goal is to reveal how these groups stay influential and continue to recruit despite being under greater scrutiny. The findings could help improve strategies for detecting and stopping criminal groups, providing useful insights for criminology and counterterrorism efforts.

lsa logoum logo