中文词频统计与词云生成

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

中文词频统计

1. 下载一长篇中文小说。

2. 从文件读取待分析文本。

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.add_word('天罡北斗阵')  #逐个添加

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

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

6. 排序

7. 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]

8. 输出词频最大TOP20，把结果存放到文件里

9. 生成词云。

 

import jieba
import struct
import os
from wordcloud import WordCloud
import matplotlib.pyplot as plt
from scipy.misc import imread

result_path = r'C:\Users\LJ\Desktop\wordcloud\result.txt'
fiction_path = r'C:\Users\LJ\Desktop\wordcloud\天龙八部.txt'
stops_path = r'C:\Users\LJ\Desktop\wordcloud\stops_chinese.txt'
userdict_path = r'C:\Users\LJ\Desktop\wordcloud\userdict\天龙八部词库.txt'
def save_result():
    # 读取小说
    with open(fiction_path, 'r', encoding='utf8') as f:
        fiction = f.read().replace('\n', '')
    #   读取停用词
    with open(stops_path, 'r', encoding='utf8') as f:
        stops = f.read().split('\n')
    #     添加用户自定义字典
    jieba.load_userdict(userdict_path)
    #  分词并发挥list
    wordlist = jieba.lcut(fiction)
    # 去除停用词
    wordlist_nostop = [word for word in wordlist if word not in stops]
    wordfrequency = {}
    # 统计词频
    for i in wordlist_nostop:
        if i not in wordfrequency:
            wordfrequency[i] = 1
        else:
            wordfrequency[i] += 1
    #   list才可排序 所以把set变为list
    paixu = list(wordfrequency.items())
    # 以value排序
    paixu.sort(key=lambda x: x[1], reverse=True)
    # Top20
    paixu = paixu[0:20]
    result = ''
    # 取key转成string
    for i in paixu:
        result = result + i[0] + ' '
    #     保存top20
    with open(result_path, 'w', encoding='utf8') as f:
        f.write(result)
def read_result():
    #   读取top20
    with open(result_path, 'r', encoding='utf8') as f:
        return f.read()
save_result()
result = read_result()
# 读取图片，
im = imread(r'C:\Users\LJ\Desktop\mask.jpg')
# 遮罩图为im
mywc = WordCloud(background_color='pink', mask=im, margin=1).generate(result)
plt.imshow(mywc)
plt.axis("off")
# 显示词云
plt.show()

运行结果：

遮罩图：

Top20:

词云：