一个完整的大作业

1.选取一个自己感兴趣的主题，我选取了搜狐新闻

登录网站：http://news.sohu.com/

 

2.在浏览器中按F12进入查看器

 

 

3.网络上爬取相关的数据，并输出结果

import requests
from bs4 import BeautifulSoup

url = 'http://news.sohu.com/'
res = requests.get(url)
res.encoding = 'UTF-8'

soup = BeautifulSoup(res.text, 'html.parser')

for news in soup.select('.list16'):
    li = news.select('li')  
    if len(li) > 0:      
        title = li[0].text       
        href = li[0].select('a')[0]['href']
        print(title, href)

 

4.进行文本分析，生成词云

from os import path 

from scipy.misc import imread   

import jieba 

import sys 

import matplotlib.pyplot as plt 

from wordcloud import WordCloud, STOPWORDS, ImageColorGenerator   

text = open('D:\\sohu.txt').read() 

wordlist = jieba.cut(text) 

wl_space_split = " ".join(wordlist) 

d = path.dirname(__file__) 

nana_coloring = imread(path.join(d, "D:\\0.jpg")) 

my_wordcloud = WordCloud( background_color = 'white',   

                            mask = nana_coloring,        

                            max_words = 3000,           

                            stopwords = STOPWORDS, 

                            max_font_size = 70,       

                            random_state = 50,            ) 

    

text_dict = {   'you': 2993,   'and': 6625,   'in': 2767,   'was': 2525,   'the': 7845,}

my_wordcloud = WordCloud().generate_from_frequencies(text_dict)

 

image_colors = ImageColorGenerator(nana_coloring) 

my_wordcloud.recolor(color_func=image_colors) 

plt.imshow(my_wordcloud)   

plt.axis("off")            

plt.show()    

my_wordcloud.to_file(path.join(d, "cloudimg.png")) 

5.结果