TOWARDS SEMANTICS: WORD SENSE DISAMBIGUATION

Word sense disambiguation (WSD) is the process of determining which sense of a polysemous word has been used in a particular context. It is perhaps one of the most difficult tasks in the field of computational linguistics. A break-through on this topic would have significant impact on numerous relevant NLP applications. For example, in machine translation, there are two main procedures which are source language understanding and target language generating. WSD can benefit the system when a polysemous word is in either the source or target language. In information retrieval, WSD can increase the quality of IR system by determining the meanings of terms (words). In Question and answering, WSD can help to extracting the correct answers for the question. In text simplification, WSD can help further reducing texts and maintain the same meaning and information. However, up to now, WSD has not been successfully applied to any of these systems. This study focuses on learning current WSD methods, such as polysemous word’s description, sense determination, and result evaluation in order to build foundation for creating a novel WSD system.