Comparison of different POS tagging techniques (n-gram, HMM and Brill's tagger) for Bangla

There are different approaches to the problem of assigning each word of a text with a parts-of-speech tag, which is known as Part-Of-Speech (POS) tagging. In this paper we compare the performance of a few POS tagging techniques for Bangla language, e.g. statistical approach (n-gram, HMM) and transformation based approach (Brill's tagger). A supervised POS tagging approach requires a large amount of annotated training corpus to tag properly. At this initial stage of POS-tagging for Bangla, we have very limited resource of annotated corpus. We tried to see which technique maximizes the performance with this limited resource. We also checked the performance for English and tried to conclude how these techniques might perform if we can manage a substantial amount of annotated corpus.