Need python code for extracting ngram from text files?And then check for similarity among all text files.

2.  extract all-and-only technical ngrams/terms/phrases (legal matter or subject matter) from the corpus.  Show
me the code you used to extract the ngrams/terms/phrases, describe the method you used in a few bullet points,
show some of the ngrams/terms/phrases, and link to the list of all ngrams/terms/phrases.

3.  for each pdf, tell me which other pdf is most similar (and for one pdf and its most similar pdf, tell me what
you consider the basis of the similarity, e.g., a list of shared ngrams/terms/phrases).

Text files are sent in gmail 

So extract ngrams and then check for similarity from a folder full of text files.Read all above instructions carefully.