Proscovia Olango (2006)
EReXS: Events and Relations Extraction for SWHi
Master's thesis, Rijksuniversiteit Groningen.
[ Paper (PDF, 428 kb) ]

Abstract

"Automatic event extraction from fulltext resources is a combination of human language technology (HLT) and semantic web technologies. It can also be done on the base of purely statistical means with minimal linguistic knowledge"[1]. This thesis introduces a semi-automated method based on the HLT approach. The method uses an existing information extraction system called ANNIE, A Nearly-New Information Extraction System (developed by Hamish Cunningham, Valentin Tablan, Diana Maynard, Kalina Bontcheva, Marin Dimitrov and others). Further text analysis is supported by WordNet and parsers that help in the automatic extraction of historical events and their relations to ob jects of the human society. Although the method is developed for fulltext resources in the eld of history, it is anticipated that it shall also be applied to e-resources in other elds for automatic extraction of historical events. The sub ject of history is well reckoned with its chronological record of true events, leading from the past to the present and even into the future.


[1] Thierry Declerck, Automatic event extraction from text on the base of linguistic and semantic annotation, DFKI Language Technology Lab. German Research Center for Artificial Intelligence GmbH, 2005