The detection of potentially illegal activity on financial discussion boards using information extraction

Owda, M and Knott, E (2012) The detection of potentially illegal activity on financial discussion boards using information extraction. In: 2nd International Conference on Cybercrime, Security and Digital Forensics, 14 May - 15 May 2012, London.

Preview

Available under License In Copyright. Download (430kB) | Preview

Abstract

flourish. This paper presents a novel prototype system for a financial share discussion detection system, to highlight potentially unlawful practices. Information is extracted from financial discussion boards, where templates hosting scenarios of known illegal activities are used to detect any potential misdemeanors. From an analysis of a single day’s trading, it was observed that of the 3000 comments extracted, 0.2% of these were deemed suspicious and required the investigation of a discussion board moderator. The man-power required to perform this task manually over the course of a year could be a prohibitive. The initial work underscores the importance and need of an automated crime detection system, using financial discussion boards as its key extraction component.