A modified technique for Word Sense Disambiguation using Lesk algorithm in Hindi language

Radhike Sawhney, Arvinder Kaur

Abstract: Word Sense Disambiguation (WSD) is a key factor in written and verbal communication of natural language processing. It is a method of selecting the appropriate sense of an ambiguous word in the given context. This paper aims at determining the correct sense of the given ambiguous word in Hindi language. A modified Lesk approach is used which uses the concept of dynamic context window. Dynamic context window is the number of left and right words of ambiguous word. The basic assumption of this approach is that the target word which has same meaning must have a common topic in its neighbourhood. Furthermore the improvement in precision shows that the proposed algorithm gives better results than the previous approaches which uses fixed size context window.