def buildNounPhraseDict(taggedReviews,
applyFn=sentenceSumSentiment):
nounPhraseDict = defaultdict(lambda: 0)
for taggedReview in taggedReviews:
for taggedSentence in taggedReview:
for np in extract_noun_phrases(taggedSentence.sentence):
nounPhraseDict[np] += applyFn(taggedSentence)
return normalize(nounPhraseDict)
def make_noun_phrase(reviews, prune):
sent_noun = []
for (words, score) in reviews:
nouns = noun_phrases_extraction.extract_noun_phrases(words)
if(prune):
token_nouns = [noun for noun in nouns if len(word_tokenize(noun))>=2 ]
else:
token_nouns = nouns
if(score > 0):
rate = "pos"
elif(score < 0):
rate = "neg"
else:
rate = "neu"
sent_noun.append((token_nouns, rate))
return sent_noun
