gensim
numpy
nltk
stop_words
langdetect
