Thank you for attending my Layer8 talk earlier this June. If you missed it or could not attend the conference, I've included the slides and references below.
Track 2: Using OSINT and NLP to Track Jihadists in Conflict Zones by Zeshan Aziz
Due to the disintermediation caused by internet platforms, it is difficult to track extremists. As expected, many of the previously useful automated heuristics for detecting extremists on social media have started to become less relevant. Extremists and their sympathizers have updated their OPSEC and TTPs to counter common detection mechanisms. While some have left mainstream sites like Facebook or Twitter, a large number – mainly those that seek to reach younger and more impressionable audiences – are still online despite clampdowns. In this talk we will present tools and techniques that OSINT analysts can borrow from the data science community to identify and discover targets of interest. With our methods, we were to find a network of extremists and foreign fighters on the ground in Syria. We tracked and analyzed the digital footprint of the late ‘Salman Belarus’ of Malhama Tactical, the infamous private military contractor for jihadis out of Syria. Our methods also found Taliban accounts and some potential members living in Europe. We discovered specific phrases and linguistical modifiers that are unique to this group and are typically used by extremists and sympathizers. Our techniques make it so your analysts need not be linguists to do this research.