Event detection using lexical chain

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 to extract its participants. It follows an unsupervised method for identifying the lexical chains from the raw sentences taken as a training data. The lexical chain constructed using Wordnet lexicon is then used for identifying event mention. The significance of the proposed system is it is the first system that applies lexical chain for event identification. The entire architecture is divided into three tasks namely, natural language pre-processing, lexical chain construction and event detection. © 2010 Springer-Verlag Berlin Heidelberg.