python情感分析实例

运行结果，代码及所用的图片，以及运行的的数据放到我的资源里面

数据爬取

import requests
from lxml import etree
import time

if __name__ == "__main__":

    fp = open('./comment.csv',mode = 'w',encoding='utf-8')
    fp .write('author\tp\tvote\n')
    url = 'https://movie.douban.com/subject/35051512/comments?start=%d&limit=20&status=P&sort=new_score'

    headers = {
        'Accept': 'text/html,application/xhtml+xml,application/xml;q=0.9,image/webp,image/apng,*/*;q=0.8,application/signed-exchange;v=b3;q=0.9',
        # 'Accept-Encoding':'gzip, deflate, br',
        'Accept-Language': 'zh-CN,zh;q=0.9,en;q=0.8',
        'Cache-Control': 'no-cache',
        'Connection': 'keep-alive',
        #'Cookie': '', #用自己浏览器上面的cookie，注释这条也可以运行
        'Host': 'movie.douban.com',
        'Pragma': 'no-cache',
        'Referer': 'https://movie.douban.com/subject/35051512/?from=playing_poster',
        'Sec-Fetch-Dest': 'document',
        'Sec-Fetch-Mode': 'navigate',
        'Sec-Fetch-Site': 'same-origin',
        'Sec-Fetch-User': '?1',
        'Upgrade-Insecure-Requests': '1',
        'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/83.0.4103.97 Safari/537.36',

    }

    for i in range(26):
        if i == 25:
            url_climb = url % (490)
        else:
            url_climb = url % (i * 20)

        response = requests.get(url=url_climb, headers=headers)

        response.encoding = 'utf-8'
        text = response.text

        html = etree.HTML(text)

        comments = html.xpath('//div[@id="comments"]/div[@class="comment-item "]')

        
        for comment in comments:
            # 作者
            author = comment.xpath('./div[@class="avatar"]/a/@title')[0]

            p = comment.xpath('.//span[@class="short"]/text()')[0]
            # 点赞
            vote = comment.xpath('.//span[@class="votes vote-count"]/text()')[0].strip()
            fp.write('%s\t%s\t%s\n'%(author,p,vote))

        print('第%d页数据保存成功'%(i+1))
        time.sleep(1)
    fp.close()

情感分析

import pandas as pd
from snownlp import SnowNLP


#该方法的作用将评论，进行情感分析
def convert(comment):
    
    snow = SnowNLP(str(comment))
    sentiments = snow.sentiments #0(消极)————1(积极)
    return sentiments
    

if __name__ == "__main__":
    data = pd.read_csv('./comment1.csv','\t')

    #获取评论数据，进行情感分析，DataFtame新增加一列
    #属性名：情感分析
    
    data['情感评分'] = data.p.apply(convert)
    
    data.sort_values(by = '情感评分',ascending=False,inplace=True)

    data.to_csv('./comment_snownlp.csv',sep = '\t',index = False)

分片提取，jieba分析词语重要性，绘制词语柱状图，和词云

import pandas as pd
import jieba
from jieba import analyse
import matplotlib.pyplot as plt
import numpy as np
import wordcloud
from PIL import Image


if __name__ == "__main__":

    data = pd.read_csv('./comment1.csv',sep='\t')
    #获取评论信息

    comments = ';'.join([str(c) for c in data['p'].tolist()])
    #使用jieba库对文本内容进行分词
    gen = jieba.cut(comments)

    words = ' '.join(gen)

    #对分好词，进行jieba分析

    tags = analyse.extract_tags(words,topK=500,withWeight=True)



    word_result = pd.DataFrame(tags,columns=['词语','重要性'])
    word_result.sort_values(by = '重要性',ascending=False,inplace=True)#从大到小排序

    #可视化，选取20个

    plt.barh(y = np.arange(0,20),width =word_result[:20]['重要性'][::-1])
    plt.ylabel('Importance')
    plt.yticks(np.arange(0,20),labels=word_result[:20]['词语'][::-1],fontproperties='KaiTi')
    plt.show()

    #词云操作
    person = np.array(Image.open('./111.jpg'))
    #将tags，jieba分词提取出数据，dict词典
    words = dict(tags)
    cloud=wordcloud.WordCloud(width=1200,height=968,font_path='./simkai.ttf',background_color='white',
                        mask=person,max_words=500,max_font_size=150,)
    word_cloud = cloud.generate_from_frequencies(words)
    plt.figure(figsize=(12,12))
    plt.imshow(word_cloud)
    plt.show()