Improving Vietnamese POS tagging by integrating a rich feature set and Support Vector Machines

Part of speech (POS) tagging is fundamental in natural language processing. So far, many methods have been applied for English and the task is well solved. However, there are few studies about this problem for Vietnamese. In this paper, we evaluate common features for English POS tagging and then propose some language specific features for Vietnamese POS tagging. Experimental results on the Vietnamese Lexicography Centerpsilas research grouppsilas corpus show that our POS tagger using this feature set trained by SVM outperforms other Vietnamese POS taggers.