Techniques for Disambiguation of Polysemy Words: A Review

Abstract

The domain of Computational Linguistics involves the key task of Word Sense Disambiguation which aims to assign a meaning to particular word in terms of the context with which it is used in a sentence. The task of assigning the semantically correct meaning to a polysemy word in almost all the languages of the world stands out to be an open problem of research with considerably low accuracies achieved. The paper presents a meticulous review of the various techniques opted for disambiguation of polysemy words in various languages-English, Hindi, Nepalese, Tamil, Kannada, Telugu, Malayalam, Sinhala and German. Also, an insight into how the various approaches-supervised (involving corpora) and Unsupervised (clustering, meta thesaurus) to solving the above problems evolved over the years to get the accuracy improved. The applications include word processing, spell checking, content analysis, translation, improved search engines.