tf-idf：sklearn中TfidfVectorizer使用

tf-idf：sklearn中TfidfVectorizer使用

import jieba

from sklearn.feature_extraction.text import TfidfVectorizer

corpus = []
data_file="./tfidf-data.txt"
with open(data_file, 'r') as f:
    for line in f:
        corpus.append(" ".join(jieba.cut(line.split(',')[0], cut_all=False)))

vectorizer = TfidfVectorizer()
tfidf = vectorizer.fit_transform(corpus)
print(tfidf.shape)

words = vectorizer.get_feature_names()
for i in range(len(corpus)):
    print('----Document %d----' % (i))
    for j in range(len(words)):
        if tfidf[i,j] > 1e-5:
              print( words[j], tfidf[i,j])