一个完整的大作业

1.选一个自己感兴趣的主题。

 我选择最近的十九大会议进行爬虫操作，爬取网站“http://cpc.people.com.cn/19th/GB/414745/414893/index.html?_zbs_baidu_dk”

2.网络上爬取相关的数据。

爬取此网页中的新闻标题，来源和时间。

import requests
from bs4 import BeautifulSoup

url = 'http://cpc.people.com.cn/19th/GB/414745/414893/index.html?_zbs_baidu_dk'
res = requests.get(url)
res.encoding='gb2312'  
soup=BeautifulSoup(res.text,'html.parser')

 

for news in soup.select('.focusBox'):
    if len(news.select('p'))>0:
        title=news.select('p')[0].text
        url=news.select('a')[0]['href']

        resd=requests.get(url)
        resd.encoding='gb2312'
        soupd=BeautifulSoup(resd.text,'html.parser')
        time=soupd.select('.sou')[0].text                      
        source=soupd.select('.sou')[0].text
        #p = soupd.select('.news_content_con')[0].text
        print(title,url,time,source)

 

3.进行文本分析，生成词云。

import requests
import jieba
from bs4 import BeautifulSoup
import re

url = 'http://cpc.people.com.cn/19th/GB/414745/414893/index.html?_zbs_baidu_dk '
res = requests.get(url)
res.encoding='gb2312'   
soup=BeautifulSoup(res.text,'html.parser')



for news in soup.select('.focusBox'):
    if len(news.select('p'))>0:
        title=news.select('p')[0].text
        url=news.select('a')[0]['href']
     

        resd=requests.get(url)
        resd.encoding='gb2312'
        soupd=BeautifulSoup(resd.text,'html.parser')
        p = soupd.select('.show_text')[0].text
        #print(p)
        break

words = jieba.lcut(p)
ls = []
counts = {}
for word in words:
    ls.append(word)
    if len(word) == 1:
        continue
    else:
        counts[word] = counts.get(word,0)+1
items = list(counts.items())
items.sort(key = lambda x:x[1], reverse = True)
for i in range(10):
    word , count = items[i]
    print ("{:<5}{:>2}".format(word,count))

from wordcloud import WordCloud
import matplotlib.pyplot as plt    
cy = WordCloud(font_path='msyh.ttc').generate(p)
plt.imshow(cy)
plt.axis("off")
plt.show()

 

4.对文本分析结果解释说明

 对文本进行分词，将词汇写入词云中让人更好地了解文本的主要内容和主题。

 

import requests
from bs4 import BeautifulSoup
from datetime import datetime
import pandas
import sqlite3



def getdetail(url):
    resd=requests.get(url)
    resd.encoding='gb2312'
    soupd=BeautifulSoup(resd.text,'html.parser')
    news={}
    news['url']=url
    news['title']=soupd.select('p')[0].text
    news['time']=soupd.select('.sou')[0].text                       
    news['source']=soupd.select('.sou')[0].text
    #news['p'] = soupd.select('.show_text')[0].text
    return(news)
    



def onepage(pageurl):
    res = requests.get(pageurl)
    res.encoding='gb2312'   
    soup=BeautifulSoup(res.text,'html.parser')
    newsls = []
    for news in soup.select('.focusBox'):
        if len(news.select('p'))>0:
            newsls.append(getdetail(news.select('a')[0]['href']))
    return(newsls)

#print(onepage('http://cpc.people.com.cn/19th/GB/414745/414893/index.html?_zbs_baidu_dk/'))
     
newstotal = []
dmurl='http://cpc.people.com.cn/19th/GB/414745/414893/index.html?_zbs_baidu_dk/'
newstotal.extend(onepage(dmurl))

res = requests.get(dmurl)
res.encoding= 'gb2312'
soup=BeautifulSoup(res.text,'html.parser')



listurl='http://cpc.people.com.cn/19th/GB/414745/414893/index.html?_zbs_baidu_dk/'
newstotal.extend(onepage(listurl))

df = pandas.DataFrame(newstotal)
print(df.head())

df.to_excel('mnews.xlsx')

with sqlite3.connect('mnewsdb.sqlite') as db:
    df.to_sql('mnewsdb8',con = db)