SVM文本分类实验过程

1、分词，打标签；

2、特征选择： 卡方检验

def chi_select():
#构建停用词表
    stopwords=[]
    with open("../hlt_stop_words.txt","r") as stopword:   
        for line in stopword: #遍历文件，一行行遍历，读取文本
            rs = line.replace('\n', '') 
            stopwords.append(rs)
#读入文本 
    x_text = 1_examples + 2_examples + ...
# 去停用词 
    x_stop=[] 
    for word in x_text:
#        word = word.split(" ")
        rs = []
        for _ in word:
            if _ not in stopwords:
                rs.append(_)
        x_stop.append(rs)

#重组
    x_final=[]
    for i in x_stop:
        x_final.append(str(i))

#建立词典
    max_document_length = max([len(x) for x in x_stop])
    #print(max_document_length)
    vocab_processor = learn.preprocessing.VocabularyProcessor(max_document_length)
    x = np.array(list(vocab_processor.fit_transform(x_final)