A Comparison of Naïve Bayesian and Nearest Neighbor Cosine Classifiers for Myanmar Word Sense Disambiguation

Abstract

This paper presents Word Sense Disambiguation for Myanmar Language. Word Sense Disambiguation (WSD) is an intermediate but an important step in Natural Language processing. WSD is defined as the task of finding the correct sense of a word in a specific context.WSD systems can help to improve the performance of statistical machine translation (MT) system. In the most used classifiers, Nearest Neighbor Cosine (NNC) model has excellent performance, and Naïve Bayesian (NB) is preferred by researchers for it is simple and useful. In this paper, we choose NNC and NB as classifiers to disambiguate ambiguous Myanmar words with part-of-speech ‘noun’, ‘verb’ and ‘adjective’. The WSD module developed here will be used as a complement to improve Myanmar-English machine translation system. As an advantage, the system can improve the accuracy of Myanmar to English language translation. We present a comparison of two methods in our experiments.