Python3.5 数据处理 --jieba + sklearn库安装及第一个示例

一，安装pip3：

#sudo apt-get install pip3

二，安装jieba：

#sudo pip3 install jieba

三，安装sklearn：

#sudo pip3 install scikit-learn

四，安装sklearn依赖(numpy,scipy)：

#sudo pip3 install numpy
#sudo pip3 install scipy

eg:国内安装时可能出现time-out错误---解决办法如下：

#sudo pip3 install -i https://pypi.tuna.tsinghua.edu.cn/simple numpy
#sudo pip3 install -i https://pypi.tuna.tsinghua.edu.cn/simple scipy

五，简单实现分词并计算TF-IDF值：

#!/usr/bin python3.5
# coding=utf-8
#import os
import jieba
import re
from sklearn.feature_extraction.text import TfidfTransformer
from sklearn.feature_extraction.text import CountVectorizer
#import collections

class Tfi(object):
    def __init__(self):
        self.stop_list = []
        pass

    def fenci(self, file):
        #list = []
        fin = open(file, ‘r‘)
        read_b = fin.read()
        fin.close()
        read_res = ‘‘.join(re.findall(u‘[a-zA-Z0-9\u4e00-\u9fa5]+‘, read_b))
        cut_res = jieba.cut(read_res, cut_all=True)
        line_res = ‘‘
        for i in cut_res:
            if i not in self.stop_list:
                line_res = line_res + i + ‘ ‘
        fout = open(‘res/‘ + file, ‘w‘)
        fout.write(line_res)
        fout.close()

    def cipin(self, fil_list):
        corpus = []
        for fil in fil_list:
            ffout = open(‘res/‘+fil, ‘r‘)
            read_r = ffout.read()
            ffout.close()
            corpus.append(read_r)
        vectorizer = CountVectorizer()
        transformer = TfidfTransformer()
        tfidf = transformer.fit_transform(vectorizer.fit_transform(corpus))
        word = vectorizer.get_feature_names()  # 所有文本的关键字
        weight = tfidf.toarray()
        for j in range(len(weight)):
            f = open(‘fes/‘+fil_list[j], ‘w‘)
            for i in range(len(word)):
                f.write(word[i]+‘  ‘+str(weight[j][i]) + ‘\n‘)
            f.close()

if __name__ == ‘__main__‘:
    first = Tfi()
    fil_list = [‘inputtext‘]
    first.fenci(‘inputtext‘)
    first.cipin(fil_list)