Comparison of Two Approaches for Part of Speech Tagging of Myanmar Language

Abstract

Part-of-Speech (POS) Tagging is the process of assigning the words with their categories that best suits the definition of the word as well as the context of the sentence in which it is used. There are different approaches to the problem of POS Tagging. In this paper, we use two approaches (supervised and unsupervised approaches), and compare the performance of these approaches for Tagging using Myanmar language. These approaches use rule based POS Tagger and the former requires a large amount of annotated training corpus to tag properly and the later does not need annotated training corpus. We tried to see which technique maximizes the performance with limited resources. By experiments, the best configuration is investigated using supervised approach with rule based part of speech tagger and the accuracy is 97.56%. Therefore, this approach has better performance than unsupervised approach with rule based part of speech tagger.