Corpus Linguistics: What can it do with Terrorism?
Dr Katerina T. Frantzi.
While the first aim for computational corpus linguistics was of course to serve linguistics, today the applications lie far beyond these borders. This work shows how corpus linguistic techniques can benefit information retrieval in the area of terrorism.
Our objective is the set of all the text announcements written by ‘17 November’, a terrorism organization being active in Greece for over 25 years. The summer of 2002 most (or all) of the ‘17N’ possible members were arrested. One of the interesting points is on the writer of each of the ‘17N’ public announcements. We apply corpus linguistics techniques on all the 84 announcements in order to extract information about linguistic and non-linguistic characteristics for each of them. A comparison of these characteristics for all the announcements can show if and to what degree some of them share any of these characteristics. This could be of help for mapping each of the announcements to its writer.
The results of this work is given purely in terms of the characteristics, that is we do not attempt to give any hints on whether an announcement was written by a specific person or not. This is a complicate subject beyond the scope of this work.
The above task could be theoretically done manually. However, it would then need a big number of human hands and a lot of time. And even then, such a text analysis done by humans could not at all guarantee the completeness and quality of the results.
Dr Katerina T. Frantzi (Greece)
Department of the Mediterranean Studies Faculty of Hellenic and Mediterranean Studies
University of the Aegean
Katerina T. Frantzi got her first degree in Informatics from the National University of Athens, Greece, 1992, and her PhD (Corpus Linguistics) from MMU in collaboration with UMIST, Manchester, England, 1998. Since 1999 she is at the Department of Mediterranean Studies, University of the Aegean, Rhodes, Greece.
(Virtual Presentation, English)