Context-based word sense disambiguation in Telugu using the statistical techniques

Abstract

The statistical technique proposed in this paper assigns a correct sense to the targeted polysemous word which has different meanings in different contexts. The methodology proposed in this paper which solves the well-known AI-Complete problem IS related to natural language processing which is called as Word Sense Disambiguation (WSD). The polysemous word may belong to anyone parts-of-speech assigned by the POS taggers. But there are some words which belong to same parts of speech but their meaning differs based on the context. Currently, the system disambiguates nouns and verbs. The system gives 100% coverage. The proposed method for word sense disambiguation gives best results while translation between different languages. A step forward in this field would have an impact on information extraction applications.