Sentiment analysis and subject extraction

# 模拟业务场景
reviews = [
    'It is an amazing movie.',
    'This is a dull movie. I would never recommend it to anyone.',
    'The cinematography is pretty great in this movie.',
    'The direction was terrible and the story was all over the place.']
sents, probs = [], []
for review in reviews:
    sample = {}
    words = review.split()
    for word in words:
        sample[word] = True
    pcls = model.classify(sample)
    print(review, '->', pcls)


输出结果：
0.735
It is an amazing movie. -> POSITIVE
This is a dull movie. I would never recommend it to anyone. -> NEGATIVE
The cinematography is pretty great in this movie. -> POSITIVE
The direction was terrible and the story was all over the place. -> NEGATIVE

 

 

'' ' 
    Subject extraction: After the word, the word cleaning, stemming, TF-IDF algorithm can be extracted core word relating to a text in order to determine the current topic text. 
            It belongs to unsupervised learning. gensim module provides a common theme extraction tool. 
    Theme extract the relevant API: 
            Import gensim.models.ldamodel gm AS 
            Import gensim.corpora AS gc 

            # lines_tokens the words that appear in the dictionary are stored in the object gc offered to do the coding for each word. 
            = line_tokens [ 'Hello', 'World', ...] 
            DIC = gc.Dictionary (line_tokens) 
            # word dictionary constructed by the bag 
            Bow = dic.doc2bow (line_tokens) 

            # Construction LDA model 
            # bow: bag of words 
            # num_topics: Classification number 
            # id2word: dictionary 
            # passes: each subject reserved for the maximum number of keywords 
            model = gm.LdaModel (bow, num_topics = n_topics, id2word = dic, passes = 25)
            The output of each category in the category # greatest contribution to 4 MeSH 
            Topics = model.print_topics (NUM_TOPICS = n_topics, NUM_WORDS =. 4) 
'' ' 

Import nltk.tokenize AS TK 
Import nltk.corpus AS NC 
Import nltk.stem.snowball SB AS 
Import gensim.models.ldamodel AS GM 
Import gensim.corpora AS GC 
DOC = [] 
with Open ( '../ ml_data / topic.txt', 'R & lt') AS F: 
    for in f.readlines Line (): 
        doc.append (Line [: -. 1]) 
the tokenizer tk.WordPunctTokenizer = () 
stopwords = nc.stopwords.words ( 'Dictionary Dictionary English') 
Signs = [ '.' '!' ',',,] 
Stemmer = sb.SnowballStemmer ( 'Dictionary Dictionary English') 
lines_tokens = [] 
for Line in DOC: 
    tokens = tokenizer.tokenize (Line.lower())
    line_tokens = []
    in token tokens for: 
        IF token and token not in stopwords not in Signs: 
            token = stemmer.stem (token) 
            line_tokens.append (token) 
    lines_tokens.append (line_tokens) 
dictionary # lines_tokens the word appear in all stores offer gc objects, do the coding for each word. 
= gc.Dictionary DIC (lines_tokens) 
# through each row to construct a list of words bags 
Bow = [] 
for line_tokens in lines_tokens: 
    Row = dic.doc2bow (line_tokens) 
    bow.append (Row) 
n_topics = 2 
# through the bag of words, the number of classifications , dictionary, the maximum number of keywords for each topic reserved build LDA model 
model = gm.LdaModel (Bow, NUM_TOPICS = n_topics, id2word = dic, passes = 25) 
# output in each category for a maximum of four theme contribution categories words 
Topics = model.print_topics (NUM_TOPICS = n_topics, NUM_WORDS =. 4) 
Print (Topics)