Conference:
The 3rd International Conference on Cybercrime, Security and Digital Forensics
Location:
Cardiff, UK
Date:
Monday, June 10, 2013
Abstract:
With the number of prosecutions for crimes on online social networks rising each year combined with the volume, velocity and variety of the data produced on these online social networks, it is almost impossible to effectively discover instances of crime or criminal behavior manually. Past work has been conducted in the area of event detection using Twitter, including TEDAS (Li, et al., 2012) and Jasmine (Watanabe, et al., 2011). This paper proposes a novel solution allowing for real-time monitoring of geo-tagged posts on Twitter using information extraction techniques to discover relevant information. This approach has had promising results, returning a number of relevant results from within the data set used in the evaluation.