POS-Tagger

Building a POS tagger to tag an unknown test corpus, given a training corpus POS tagger has been developed using Hidden Markov Model to learn from a tagged corpus. The transition and emission probabilities are generated and are used to predict tags for an unknown corpus of the same language. The model can be extended to tag any language data, given the corresponding training set. The tagging accuracy of the model for English language is as high as 86%.