source code for sentiment analysis of product/movie reviews using SentiWordNet and NLTK.

SentiWordNet is available at http://sentiwordnet.isti.cnr.it/ NLTK is available at http://nltk.org/

The approach is based on part of speech tagging and extraction of useful bigram and trigram patterns containing adjectives and adverbs. Implemented HMM tagger using tagged corpus (available as part of NLTK) and Viterbi decoding algorithm.