Word Sense Disambiguation by using Naïve Bayesian Classification

Abstract

Natural Language Processing has been developed to allow human-machine communication to take place in a natural-language. Word Sense Disambiguation (WSD) is regarded as one of the most interesting and longest-standing problems in NLP. Several methodological issues come up with the context of WSD. These are supervised vs. unsupervised WSD approaches. Supervised WSD approaches have obtained better results than unsupervised WSD approaches. Naïve Bayesian WSD approach is one of the best supervised WSD approaches. This paper presents a corpus-based approach that uses Naïve Bayesian Classification to disambiguate ambiguous words with part-of-speech ‘noun’, which uses topical feature that represent cooccurring words in bag-of-word feature. This system also uses Senseval-3 corpus as a training data for Naïve Bayesian Classification, and access Word Net for retrieving meaning of the resulted senses. This system tokenizes and tags part-of-speech to each word of input sentence, collect target words, disambiguate target words and output correct sense and meaning for each target word.