def sentiment(): if query.count() > 0: articles = query.paginate_article(1, 100) for article in articles: if article.title: s = SnowNLP(article.title) article.feeling_factor = s.sentiments else: article.feeling_factor = -2 article.save() logger.info("Sentiment SUCCEED ARTICLE<%s>" % article.uuid) else: time.sleep(10)
[(a_count[x]**2) for x in xrange(len(union))]) denominator2 = reduce(lambda x, y: x + y, [(b_count[x]**2) for x in xrange(len(union))]) value = numerator / (math.sqrt(denominator1) * math.sqrt(denominator2)) return value if __name__ == '__main__': """ test 1 """ # compare(jieba.cut(u"我喜欢看电视,不喜欢看电影"), jieba.cut(u"我不喜欢看电视,也不喜欢看电影")) """ tese from mysql """ title1 = u"中国铝业巨亏162亿元成亏损王" articles = query.paginate_article(1, 2000) for article in articles: if not title1 or not article.title: continue a = jieba.cut(title1) b = jieba.cut(article.title) value = compare(a, b) if value > 0.2: print title1.encode("utf-8"), article.title.encode("utf-8"), value
denominator1 = reduce(lambda x, y: x + y, [(a_count[x] ** 2) for x in xrange(len(union))]) denominator2 = reduce(lambda x, y: x + y, [(b_count[x] ** 2) for x in xrange(len(union))]) value = numerator / (math.sqrt(denominator1) * math.sqrt(denominator2)) return value if __name__ == '__main__': """ test 1 """ # compare(jieba.cut(u"我喜欢看电视,不喜欢看电影"), jieba.cut(u"我不喜欢看电视,也不喜欢看电影")) """ tese from mysql """ title1 = u"中国铝业巨亏162亿元成亏损王" articles = query.paginate_article(1, 2000) for article in articles: if not title1 or not article.title: continue a = jieba.cut(title1) b = jieba.cut(article.title) value = compare(a, b) if value > 0.2: print title1.encode("utf-8"), article.title.encode("utf-8"), value