中文词频统计与词云生成

作业来源：https://edu.cnblogs.com/campus/gzcc/GZCC-16SE1/homework/2822

任务：

1. 下载一长篇中文小说

2. 从文件读取待分析文本

novel = open(r'E:三体.txt', 'r', encoding='utf8').read()

3. 安装并使用jieba进行中文分词

pip install jieba

import jieba

jieba.lcut(text)

4. 更新词库，加入所分析对象的专业词汇

jieba.load_userdict(word_dict)  #词库文本文件

jieba.load_userdict(r'E:三体词库.txt')

参考词库下载地址：https://pinyin.sogou.com/dict/

转换代码：scel_to_text

5. 生成词频统计、排序

wordSet = set(tokens)

wordDict = {}
for w in wordSet:
    if len(w) > 1:
        wordDict[w] = tokens.count(w)
wordList = list(wordDict.items())
wordList.sort(key=lambda x: x[1], reverse=True)

6. 排除语法型词汇，代词、冠词、连词等停用词

stops

tokens=[token for token in wordsls if token not in stops]

with open(r'E:stops_chinese.txt', 'r', encoding='utf8') as f:
    stops = f.read().split('
')
tokens = [token for token in cutText if token not in stops]

7. 输出词频最大TOP25，把结果存放到文件里

for i in range(25):
    print(wordList[i])

8. 生成词云

install wordcloud：

 

相关代码：

pd.DataFrame(data=wordList).to_csv('E:\三体词频统计.csv', encoding='utf8')
cut_text = "".join(tokens)
im = imread(r'E:	ree.jpg')
mywc = WordCloud(background_color='white', mask=im, margin=2).generate(cut_text)
plt.imshow(mywc)
plt.axis("off")
plt.show()

9 完整代码：

import jieba
import pandas as pd
from wordcloud import WordCloud
import matplotlib.pyplot as plt
from scipy.misc import imread

novel = open(r'E:三体.txt', 'r', encoding='utf8').read()
jieba.load_userdict(r'E:三体词库.txt')
cutText = jieba.lcut(novel)

with open(r'E:stops_chinese.txt', 'r', encoding='utf8') as f:
    stops = f.read().split('
')
tokens = [token for token in cutText if token not in stops]
wordSet = set(tokens)

wordDict = {}
for w in wordSet:
    if len(w) > 1:
        wordDict[w] = tokens.count(w)
wordList = list(wordDict.items())
wordList.sort(key=lambda x: x[1], reverse=True)

for i in range(25):
    print(wordList[i])


pd.DataFrame(data=wordList).to_csv('E:\三体词频统计.csv', encoding='utf8')
cut_text = "".join(tokens)
im = imread(r'E:	ree.jpg')
mywc = WordCloud(background_color='white', mask=im, margin=2).generate(cut_text)
plt.imshow(mywc)
plt.axis("off")
plt.show()

运行结果：