py中文词频统计

# -*- coding: UTF-8 -*-
# -*- coder: mzp -*-

import jieba
import re
with open('test.txt', 'r', encoding='UTF-8') as story:
    test = story.read()
miss_word = "了|他|说|我|你|就|着|又|的|在|是|有|把|到|也|不|都|她|这|便|去|们|还|但|一个|和|却|里|来|要|没|很|\"" \
            "|那|么|一|将|呢|起|于|上|只|得|而|而且|对|所以|见|些|才|从|过|被|并|时|且|给|道|虽然|可以|出"
test = re.sub("[\s+\.\!\/_\",$%^*+—()?【】“《；》”！\-：，。？、~@#￥%……&*（）]+", "", test)
test = re.sub(miss_word + '+', "", test)
words = list(jieba.cut(test))
# test =''.join(re.findall(u'[\u4e00-\u9fff]+', test))  # 简单粗暴的中文筛选
key_words = {}
for i in set(words):  # 统计出词频
    key_words[i] = words.count(i)
sort_word = sorted(key_words.items(), key = lambda d:d[1], reverse = True)  # 排序
for i in range(20):  # 输出
    print(sort_word[i])