Tracula by mikolajb · Pull Request #3 · recap/pumpkin

mikolajb · 2014-12-03T20:05:02Z

New filters:

language detection using NLTK (supports danish, dutch, english, finnish, french, german, hungarian, italian, norwegian, portuguese, russian, spanish, swedish and turkish)
named entity extractor using ne_chunk method from NLTK
topic detection (basing on words statistics in movie_review and reuters corpus from NLTK)

- language detection using NLTK - named entity extractor using ne_chunk method from NLTK - topic detection

to use with option "--gonzales"

mikolajb added 5 commits December 3, 2014 17:09

New filters:

4943ff4

- language detection using NLTK - named entity extractor using ne_chunk method from NLTK - topic detection

dependencies of seeds removed - content added to seed's files

1977a78

fixed after extensive testing

3934ca5

consumes one tweet per message (instead of a list)

c958ee4

to use with option "--gonzales"

sentiment analyses included in nlp workflow

57d65e1

Provide feedback