Part-of-Speech Tagging Using A Variable Context Markov Model

We present anew approach to disambiguating syntactically ambiguous words in context, based on Variable Context Markov Models. In contrast to fixed-length Markov models, which predict based on fixed-length histories, Variable Context Markov models dynamically adapt their history length based on the training data, and hence may use fewer parameters. In a test of a VMM based tagger on the Brown corpus, 95.71% of tokens are correctly classified.