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

1. 下载一长篇中文小说。

2. 从文件读取待分析文本。

 f = open("红楼梦.txt", "r", encoding='gb18030')
    novel = f.read()
    f.close()

3. 安装并使用jieba进行中文分词。

　　

4. 更新词库，加入所分析对象的专业词汇

  jieba.add_word('林姑娘')
    jieba.load_userdict(r'红楼梦词库.txt'

5. 生成词频统计

for i in tokens:
        wordsCount[i] = tokens.count(i)

6. 排序

top = list(wordsCount.items())
top.sort(key=lambda x: x[1], reverse=True)

7.排除语法型词汇，代词、冠词、连词等停用词

f = open("stop_chinese.txt", "r", encoding='utf-8')
stops = f.read().split()
f.close()
tokens = [token for token in novel if token not in stops]

8. 输出词频最大TOP20，把结果存放到文件里

top.sort(key=lambda x: x[1], reverse=True)
pd.DataFrame(data=top[0:20]).to_csv('top_chinese20.csv', encoding='utf-8')

9. 生成词云　

　

txt = open('top_chinese20.csv','r',encoding='utf-8').read()
 wordlist = jieba.lcut(txt)

 wl_split  = ''.join(wordlist)
 backgroud_Image = plt.imread('background.jpg')

 mymc = WordCloud(background_color='white',mask=backgroud_Image,
                  margin=2,max_words=20,max_font_size=150,random_state=30).generate(wl_split)
  
 plt.imshow(mymc)
 plt.axis("off")
 plt.show()
 mymc.to_file(r'WordCloud.png')