Lexicalized HMM-based Part-of-Speech Tagger for Myanmar Language

Abstract

Part-of-speech (POS) tagging is the process of assigning the part-of-speech tag or other lexical class marker to each and every word in a sentence. In many Natural Language Processing (NLP) applications such as word sense disambiguation, information retrieval, information processing, parsing, question answering, and machine translation, POS tagging is considered as the one of the basic necessary tools. Identifying the ambiguities in lexical items is the challenging objective in the process of developing an efficient and accurate POS Tagger. This paper proposes the developments for POS-tagger and POS-tagset of Myanmar language, which is very essential computational linguistic tool needed for many natural language processing (NLP) applications. Since most previous works for HMM-based tagging consider only part-of-speech information in contexts, their models cannot utilize lexical information which is crucial for resolving some morphological ambiguity. In this paper, a simple method to build Lexicalized Hidden Markov Models (L-HMMs) is introduced for improving the precision of part-of-speech tagging in Myanmar language. In experiments, lexicalized models achieve higher accuracy than non-lexicalized models.