comparison.py

import extractScriptEntities
import extractSummaryEntities 
import spacy
import numpy as np
import gensim
from sklearn import preprocessing
from gensim.models import Word2Vec

"""
import gensim
from nltk.sentiment.vader import SentimentIntensityAnalyzer
from nltk.stem import WordNetLemmatizer
from nltk.tokenize import sent_tokenize, word_tokenize
from nltk.stem.snowball import SnowballStemmer
"""
#nlp = spacy.load('en')

"""
scenes = extractScriptEntities.extractScriptEntities('script.xml')
summarySentences = extractSummaryEntities.extractSummaryEntities('wikiplot.txt')

lemmatizer = WordNetLemmatizer()
stemmer = SnowballStemmer("english")

words = word_tokenize('He is incarcerated. He is in incarceration')

lemmas = [lemmatizer.lemmatize(word, pos = 'v') for word in words]
stems = [stemmer.stem(word) for word in words]

print (lemmas)
print(stems)
"""
"""
doc1 = nlp('dog')
doc2 = nlp('murder')
print(doc1, doc2)

print(doc1.similarity(doc2))

#print(doc1.vector)
print((np.dot(doc1.vector/np.linalg.norm(doc1.vector), doc2.vector/np.linalg.norm(doc2.vector))))
"""

script = open('script.txt')
sentences = [sentence for sentence in script]
splitSentences = [sentence.split() for sentence in sentences]


model = gensim.models.KeyedVectors.load_word2vec_format('GoogleNews-vectors-negative300.bin', binary = True)
#model = Word2Vec(splitSentences, min_count = 1)
#word_vectors = model.wv
#print('American' in word_vectors.vocab)

#model.train(splitSentences,total_examples = len(splitSentences), epochs = 1)
print(2)
try:
	print(model.similarity('need', 'want'))
	print(model.similarity('cat', 'dog'))
	print(model.similarity('murder', 'kill'))
	print(model.similarity('gun', 'kill'))
	print(model.similarity('interview', 'interviewing'))
	print(model.similarity('cat', 'murder'))
except:
	print(3)