Tuesday, May 23, 2006

Arabic Extremist Group Forum Messages' Characteristics

Ever wondered what's the font size of a terrorist forum posting? These guys are really deep into using AI for gathering intelligence on various Cyberterrorism threats, and as you can see they neatly visualize their findings. "Applying Authorship Analysis to Extremist-Group Web Forum Messages" by Ahmed Abbasi and Hsinchun Chen, University of Arizona seem to have found a way, or at least patters of ongoing terrorist communication, and of course propaganda online. What they did was :



"To explore these problems, we modified an existing framework for analyzing online authorship and applied it to Arabic and English Web forum messagesassociated with known extremist groups. We developed a special multilingual model—the set of algorithms and related features—to identify Arabic messages, gearing this model toward the language’s unique characteristics. Furthermore, we incorporated a complex message extraction component to allow the use of a more comprehensive set of features tailored specifically toward online messages. A series of experiments evaluating the models indicated a high level of success in identifying communication patterns."



Social network analysis has a lot of potential, and with data mining it seems to be the perfect match for the recent trouble with NSA's domestic spying program. DearNSA.com and the Patriot Search are aiming to solve the problem for both parties -- efficiently.



There's a lot of propaganda chat going on online all the time, and among the very few limitations that bother me about such web aggregation of open source information are the use of steganography, or plain-simple Dark Web (closed for crawlers with basic/sophisticated authentication in place) communication -- remember there's a lot of noise to sort out through as well.