def flesch_index(text):
sentences = sent_tokenize(text)
total_sentences = len(sentences)
total_words = np.sum([word_count(s) for s in sentences])
total_syllables = np.sum([ np.sum([ syllable_count(w) for w in word_tokenize(s) ]) \
for s in sentences ])
return 0.39*(total_words/total_sentences) + \
11.8*(total_syllables/total_words) - 15.59
def find_wiki_page(syllable_constraint=5):
while True:
tmp = wikipedia.random()
if syllable_count.syllable_count(tmp) == syllable_constraint\
and not forbidden_phrases_in_result(tmp):
return tmp
def hard_word_count(sent):
return len([ w for w in word_tokenize(sent) \
if syllable_count(wnl.lemmatize(w, pos='v'))>=3 ])