Title: Event detection using trigger chain

Authors: S. Sangeetha; R.S. Thakur; Michael Arock

Addresses: Department of Computer Applications, National Institute of Technology, Tiruchirapalli – 620015, Tamilnadu, India. ' Department of Computer Applications, Maulana Azad National Institute of Technology, Bhopal – 462051, Madhya Pradesh, India. ' Department of Computer Applications, National Institute of Technology, Tiruchirapalli – 620015, Tamilnadu, India

Abstract: This paper describes a new architecture for event detection from text documents. The proposed system correctly identifies the sentences that describe an event of interest, using trigger chain to extract its participants. It exploits supervised method for identifying the lexical chains from the raw sentences adopted as a training data. Lexical chain holds the set of semantically related words of a document from which it was obtained. The proposed system learns lexical chain, prepositions, and types of named entities from the training data to construct a trigger chain. It identifies events using this trigger chain. The entire architecture is divided into three tasks namely; natural language pre-processing, trigger chain construction and event identification.

Keywords: event detection; event identification; trigger chain; text documents; lexical chains; natural language pre-processing; semantics; information extraction; information retrieval.

DOI: 10.1504/IJKEDM.2012.044708

International Journal of Knowledge Engineering and Data Mining, 2012 Vol.2 No.1, pp.76 - 88

Published online: 02 Sep 2014 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article