This project uses latent models to align different tutorials for the same task. The model should also discriminate between tutorials of different tasks.
HMM code has been obtained from https://github.com/guyz/HMM
To run the code:
In one terminal - cd stanford-corenlp-python python corenlp.py
Wait for Stanford CoreNLP to start up. Then in another terminal, cd hmm/recipe/bigram python TestRecipe.py
Other language model based HMMs can be run in the same manner from folders hmm/recipe/background, hmm/recipe/verb_noun_unigram and hmm/recipe/verb_noun_tuple.