This file, marked "confidential", describes development of an EU-funded intelligence gathering system ("INDECT work package 4") designed to comb webblogs, chat sites, newsreports, and social-networking sites to inorder to build up automatic dossiers on individuals, organizations and their relationships.
"The aim of work package 4 (WP4) is the development of key technologies that facilitate the building of an intelligence gathering system by combining and extending the current-state-ofthe- art methods in Natural Language Processing (NLP). One of the goals of WP4 is to propose NLP and machine learning methods that learn relationships between people and organizations through websites and social networks. Key requirements for the development of such methods are: (1) the identification of entities, their relationships and the events in which they participate, and (2) the labelling of the entities, relationships and events in a corpus that will be used as a means both for developing the methods."
The file was (accidently?) released at http://indect-project.eu/files/deliverables/public/INDECT_Deliverable_4.1_v20090630a.pdf/at_download/file