Marathi parts-of-speech tagger using supervised learning

Abstract

In this paper, we present a parts-of-speech tagger for inflectional and derivational morphologically rich language Marathi. Marathi is spoken by the native people of Maharashtra. The general approach used for the development of tagger is statistical-based hidden Markov model (HMM). We establish a methodology of parts-of-speech (POS) tagging for Marathi using HMM. The main concept of HMM is to calculate probabilities to determine which is the best sequence of tags that correspond to observation sequence of words. In this paper, we show the development of the tagger. Moreover, we have also shown the evaluation done.