util.py 文件源码

def preprocess_per_news(news):

        content = jieba.posseg.cut(news)  # ??????
        content = filter(lambda x: hasMeaningfulWords(x), content)  # ???????????
        content = [i.word for i in content]
        content = filter(lambda x: len(x) > 1, content)  # ?????
        stopword_file = "/Users/luoyi/Scala/OnlineRS/com/Recsys_engine/data/stop_word.txt"
        stopw = [line.strip().decode('utf-8') for line in open(stopword_file).readlines()]

        parsed = set(content) - set(stopw)
        return ' '.join(parsed)