腾讯弹幕数据分析实战

以你之姓@ 2022-12-29 12:50 256阅读 0赞

### 腾讯弹幕数据分析实战 ###

*  通用爬虫代码：
 *  令人心动的offer2可视化分析
 *   *  批量导入数据并合并
     *  数据读取
     *  数据处理及清洗
     *   *  重命名
         *  过滤字段
         *  时间格式转换
         *  机械压缩函数处理comment
         *  会员等级打标
     *  数据分析
     *   *  1.各期弹幕数量对比
         *  2.谁是弹幕发射机
         *  3.会员等级分布
         *  词云图讨论
         *  4.人物被提及
         *  情感分析
 *  数据集：
 *  总结

# 通用爬虫代码： #

''' Date: 2020-12-21 23:27:59 LastEditTime: 2020-12-22 16:36:43 '''
    import requests
    import json
    import time
    import pandas as pd
      
    
    def get_danmu_all_page(target_id, vid,filename):
        df = pd.DataFrame()
        for page in range(15, 20000, 30):  #亲测有效
            headers = { 'User-Agent': 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_14_6) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/83.0.4103.116 Safari/537.36'}
            url = 'https://mfm.video.qq.com/danmu?otype=json&timestamp={0}&target_id={1}vid{2}&count=80'.format(page,target_id,vid)
            print("正在提取第" + str(page) + "页")
            html = requests.get(url,headers = headers)
            bs = json.loads(html.text,strict = False)  #strict参数解决部分内容json格式解析报错
            time.sleep(0.5)
            #遍历获取目标字段
            try:
                for i in bs['comments']:
                    opername = i['opername']
                    content = i['content']  #弹幕
                    upcount = i['upcount']  #点赞数
                    user_degree =i['uservip_degree'] #会员等级
                    timepoint = i['timepoint']  #发布时间
                    comment_id = i['commentid']  #弹幕id
                    cache = pd.DataFrame({ '用户id':[opername],'弹幕':[content],'会员等级':[user_degree],'发布时间':[timepoint],'弹幕点赞':[upcount],'弹幕id':[comment_id]})
                    df = pd.concat([df,cache])
            except Exception as e:
                break
        # df.to_csv(f'{filename}.csv',encoding = 'utf-8')
        return df
    #自己按需添加
    target_id = ['6130942571%26','6164313448%26','6194952391%26','6227063464%26']
    vid = ['%3Dt0034o74jpr','%3Dr00346rvwyq','%3Dd0035rctvoh','%3Db0035j0tgo0']
    filename = ['面试篇','第1期','第2期','第3期']
    df = pd.DataFrame()
    for i in range(len(vid)):
        df = get_danmu_all_page(target_id=target_id[i], vid=vid[i],filename=filename[i])#df{}.format(data_dm[i])
        df.insert(0, '所属期数', filename[i])
        df.to_csv(f'{filename[i]}.csv',index=False)

# 令人心动的offer2可视化分析 #

import pandas as pd
    import numpy as np
    import pyecharts.options as opts
    from pyecharts.charts import *
    from pyecharts.globals import ThemeType

## 批量导入数据并合并 ##

# path = 'D:/Pandas/csv/'
    # #文件目录路径
    # df_all = pd.DataFrame()
    # for i in os.listdir(path):
    # df_one = pd.read_csv(path+f'{i}', engine='python', encoding='utf-8') 
    # df_all = df_all.append(df_one, ignore_index=False)#ignore_index：默认值为False，如果为True则不使用index标签
    # df_all.shape
    # #输出
    # df_all.to_excel('offer.xlsx',index=False)

## 数据读取 ##

df = pd.read_excel('offer.xlsx')
    df[:10]

<table> 
 <thead> 
  <tr> 
   <th></th> 
   <th>所属期数</th> 
   <th>用户id</th> 
   <th>弹幕</th> 
   <th>会员等级</th> 
   <th>发布时间</th> 
   <th>弹幕点赞</th> 
   <th>弹幕id</th> 
  </tr> 
 </thead> 
 <tbody> 
  <tr> 
   <th>0</th> 
   <td>第1期</td> 
   <td>NaN</td> 
   <td>47，第一来了</td> 
   <td>4</td> 
   <td>5</td> 
   <td>23</td> 
   <td>6732257356663684096</td> 
  </tr> 
  <tr> 
   <th>1</th> 
   <td>第1期</td> 
   <td>NaN</td> 
   <td>第一</td> 
   <td>1</td> 
   <td>8</td> 
   <td>5</td> 
   <td>6732257393620750336</td> 
  </tr> 
  <tr> 
   <th>2</th> 
   <td>第1期</td> 
   <td>NaN</td> 
   <td>哈哈哈，我还以为我是第一个呢</td> 
   <td>1</td> 
   <td>7</td> 
   <td>32</td> 
   <td>6732257548063721472</td> 
  </tr> 
  <tr> 
   <th>3</th> 
   <td>第1期</td> 
   <td>NaN</td> 
   <td>YEYEYEY</td> 
   <td>0</td> 
   <td>8</td> 
   <td>16</td> 
   <td>6732257672952315904</td> 
  </tr> 
  <tr> 
   <th>4</th> 
   <td>第1期</td> 
   <td>NaN</td> 
   <td>来了来了，</td> 
   <td>6</td> 
   <td>6</td> 
   <td>18</td> 
   <td>6732258003787425792</td> 
  </tr> 
  <tr> 
   <th>5</th> 
   <td>第1期</td> 
   <td>叶湘伦</td> 
   <td>来了</td> 
   <td>0</td> 
   <td>8</td> 
   <td>0</td> 
   <td>6732258629720189952</td> 
  </tr> 
  <tr> 
   <th>6</th> 
   <td>第1期</td> 
   <td>NaN</td> 
   <td>来了来了</td> 
   <td>0</td> 
   <td>8</td> 
   <td>2</td> 
   <td>6732258673319944192</td> 
  </tr> 
  <tr> 
   <th>7</th> 
   <td>第1期</td> 
   <td>海浪</td> 
   <td>来啦</td> 
   <td>0</td> 
   <td>8</td> 
   <td>0</td> 
   <td>6732258806970479616</td> 
  </tr> 
  <tr> 
   <th>8</th> 
   <td>第1期</td> 
   <td>NaN</td> 
   <td>来咯</td> 
   <td>0</td> 
   <td>5</td> 
   <td>4</td> 
   <td>6732258885488543744</td> 
  </tr> 
  <tr> 
   <th>9</th> 
   <td>第1期</td> 
   <td>熙崽是神仙</td> 
   <td>第一！</td> 
   <td>0</td> 
   <td>7</td> 
   <td>9</td> 
   <td>6732260786762655744</td> 
  </tr> 
 </tbody> 
</table>

## 数据处理及清洗 ##

### 重命名 ###

所属期数 episodes 用户id user 弹幕 comment 会员等级 grade 发布时间 date 弹幕点赞 likecounts 弹幕id dmid

df.rename(columns={ '所属期数':'episodes','用户id':'user','弹幕':'comment','会员等级':'grade','发布时间':'date','弹幕点赞':'likecounts','弹幕id':'dmid'},inplace=True)
    df.head()

<table> 
 <thead> 
  <tr> 
   <th></th> 
   <th>episodes</th> 
   <th>user</th> 
   <th>comment</th> 
   <th>grade</th> 
   <th>date</th> 
   <th>likecounts</th> 
   <th>dmid</th> 
  </tr> 
 </thead> 
 <tbody> 
  <tr> 
   <th>0</th> 
   <td>第1期</td> 
   <td>NaN</td> 
   <td>47，第一来了</td> 
   <td>4</td> 
   <td>5</td> 
   <td>23</td> 
   <td>6732257356663684096</td> 
  </tr> 
  <tr> 
   <th>1</th> 
   <td>第1期</td> 
   <td>NaN</td> 
   <td>第一</td> 
   <td>1</td> 
   <td>8</td> 
   <td>5</td> 
   <td>6732257393620750336</td> 
  </tr> 
  <tr> 
   <th>2</th> 
   <td>第1期</td> 
   <td>NaN</td> 
   <td>哈哈哈，我还以为我是第一个呢</td> 
   <td>1</td> 
   <td>7</td> 
   <td>32</td> 
   <td>6732257548063721472</td> 
  </tr> 
  <tr> 
   <th>3</th> 
   <td>第1期</td> 
   <td>NaN</td> 
   <td>YEYEYEY</td> 
   <td>0</td> 
   <td>8</td> 
   <td>16</td> 
   <td>6732257672952315904</td> 
  </tr> 
  <tr> 
   <th>4</th> 
   <td>第1期</td> 
   <td>NaN</td> 
   <td>来了来了，</td> 
   <td>6</td> 
   <td>6</td> 
   <td>18</td> 
   <td>6732258003787425792</td> 
  </tr> 
 </tbody> 
</table>

df.info()

<class 'pandas.core.frame.DataFrame'>
    RangeIndex: 70798 entries, 0 to 70797
    Data columns (total 7 columns):
     #   Column      Non-Null Count  Dtype 
    ---  ------      --------------  ----- 
     0   episodes    70798 non-null  object
     1   user        26409 non-null  object
     2   comment     70797 non-null  object
     3   grade       70798 non-null  int64 
     4   date        70798 non-null  int64 
     5   likecounts  70798 non-null  int64 
     6   dmid        70798 non-null  int64 
    dtypes: int64(4), object(3)
    memory usage: 3.8+ MB

### 过滤字段 ###

#对用户进行空值补充
    df['user'] = df['user'].fillna('未知用户')
    #对弹幕进行re匹配处理
    df['comment'] = df['comment'].str.extract(r"([\u4e00-\u9fa5]+)") #提取中文内容
    df = df.dropna()  #纯表情弹幕直接删除
    #提取分析字段
    df = df.iloc[:,:-1]
    df.head()

<table> 
 <thead> 
  <tr> 
   <th></th> 
   <th>episodes</th> 
   <th>user</th> 
   <th>comment</th> 
   <th>grade</th> 
   <th>date</th> 
   <th>likecounts</th> 
  </tr> 
 </thead> 
 <tbody> 
  <tr> 
   <th>0</th> 
   <td>第1期</td> 
   <td>未知用户</td> 
   <td>第一来了</td> 
   <td>4</td> 
   <td>5</td> 
   <td>23</td> 
  </tr> 
  <tr> 
   <th>1</th> 
   <td>第1期</td> 
   <td>未知用户</td> 
   <td>第一</td> 
   <td>1</td> 
   <td>8</td> 
   <td>5</td> 
  </tr> 
  <tr> 
   <th>2</th> 
   <td>第1期</td> 
   <td>未知用户</td> 
   <td>哈哈哈</td> 
   <td>1</td> 
   <td>7</td> 
   <td>32</td> 
  </tr> 
  <tr> 
   <th>4</th> 
   <td>第1期</td> 
   <td>未知用户</td> 
   <td>来了来了</td> 
   <td>6</td> 
   <td>6</td> 
   <td>18</td> 
  </tr> 
  <tr> 
   <th>5</th> 
   <td>第1期</td> 
   <td>叶湘伦</td> 
   <td>来了</td> 
   <td>0</td> 
   <td>8</td> 
   <td>0</td> 
  </tr> 
 </tbody> 
</table>

### 时间格式转换 ###

def time_change(seconds):
        m, s = divmod(seconds, 60)
        h, m = divmod(m, 60)
        ss_time = "%d:%02d:%02d" % (h, m, s)
        # print(ss_time)
        return ss_time
    # time_change(seconds=8888)
    #将time_change函数应用于date字段：
    df["date"] = df["date"].apply(time_change)
    #设置为需要的时间格式
    df['date'] = pd.to_datetime(df['date'])
    df['date'] = df['date'].apply(lambda x : x.strftime('%H:%M:%S'))

### 机械压缩函数处理comment ###

#定义机械压缩函数
    def yasuo(st):
        for i in range(1,int(len(st)/2)+1):
            for j in range(len(st)):
                if st[j:j+i] == st[j+i:j+2*i]:
                    k = j + i
                    while st[k:k+i] == st[k+i:k+2*i] and k<len(st):   
                        k = k + i
                    st = st[:j] + st[k:]    
        return st
    # yasuo(st='')
    #调用机械压缩函数
    df["comment"] = df["comment"].astype("str").apply(yasuo)

### 会员等级打标 ###

df.grade.value_counts().index.tolist()

[0, 3, 1, 4, 2, 5, 6, 7, 8]

df['grade'] = 'v'+df['grade'].astype('str')#['v'+i for i in df['grade']]

df.sample(2)

<table> 
 <thead> 
  <tr> 
   <th></th> 
   <th>episodes</th> 
   <th>user</th> 
   <th>comment</th> 
   <th>grade</th> 
   <th>date</th> 
   <th>likecounts</th> 
  </tr> 
 </thead> 
 <tbody> 
  <tr> 
   <th>9235</th> 
   <td>第1期</td> 
   <td>未知用户</td> 
   <td>王颖飞好像一个演员</td> 
   <td>v0</td> 
   <td>00:27:42</td> 
   <td>5</td> 
  </tr> 
  <tr> 
   <th>57421</th> 
   <td>面试篇</td> 
   <td>zxz</td> 
   <td>彩虹袜</td> 
   <td>v3</td> 
   <td>00:13:07</td> 
   <td>12</td> 
  </tr> 
 </tbody> 
</table>

## 数据分析 ##

#绘图通用函数
    def get_pyechart(x,y,chart,title,size,pos,theme):
        if chart == 'bar':
            c = (
                Bar(init_opts=opts.InitOpts(theme=theme))
                .add_xaxis(x)
                .add_yaxis("",y) 
                .set_series_opts(label_opts=opts.LabelOpts(font_size=size,position=pos))
                )
        elif chart == 'barh':
            c = (
                Bar(init_opts=opts.InitOpts(theme=theme))
                .add_xaxis(x)
                .add_yaxis("",y).reversal_axis() #X轴与y轴调换顺序
                .set_series_opts(label_opts=opts.LabelOpts(font_size=size,position=pos))
                )
        elif chart == 'pie':
            c = (
                Pie(init_opts=opts.InitOpts(theme=theme))
                .add("", list(zip(x,y)))
                .set_series_opts(label_opts=opts.LabelOpts(formatter="等级{b}占比:{d}%",font_size=size))
                )
        elif chart == 'line':
            c = (
                Line(init_opts=opts.InitOpts(theme=theme))
                .add_xaxis(x)
                .add_yaxis('情感倾向',y, is_smooth=True,is_connect_nones=True,areastyle_opts=opts.AreaStyleOpts(opacity=0.5))
                .set_global_opts(title_opts=opts.TitleOpts(title=title,subtitle="数据来源：腾讯视频",pos_left = 'left'))
                )
        c.set_global_opts(title_opts=opts.TitleOpts(title=title,subtitle="数据来源：腾讯视频",pos_left = 'left'),
                            xaxis_opts=opts.AxisOpts(axislabel_opts=opts.LabelOpts(font_size=13)), #更改横坐标字体大小
                            yaxis_opts=opts.AxisOpts(axislabel_opts=opts.LabelOpts(font_size=13)), #更改纵坐标字体大小
                            )
        return c.render_notebook()

### 1.各期弹幕数量对比 ###

df_e = df['episodes'].value_counts()
    df_e

面试篇    18603
    第3期    18107
    第1期    17551
    第2期    15844
    Name: episodes, dtype: int64

x = df_e.index.tolist()
    y = df_e.values.tolist()
    df_e = get_pyechart(x=x,y=y,chart='bar',title='各期弹幕数量',size=16,pos='top',theme=ThemeType.DARK)
    df_e

![在这里插入图片描述][watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3FxXzQ0NzYwOTEy_size_16_color_FFFFFF_t_70]

### 2.谁是弹幕发射机 ###

df_u = df['user'].value_counts()[1:10].sort_values(ascending=True)
    df_u

神马不是浮云     78
    .          79
    圣雪天使       80
    。          89
    白龙吟        92
    为时不晚i      93
    momo       93
    ベ☆小强呐     103
    想太多de猫    135
    Name: user, dtype: int64

x = df_u.index.tolist()
    y = df_u.values.tolist()
    df_u = get_pyechart(x=x,y=y,chart='barh',title='弹幕发送数量TOP10',size=16,pos='right',theme=ThemeType.DARK)
    df_u

![在这里插入图片描述][watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3FxXzQ0NzYwOTEy_size_16_color_FFFFFF_t_70 1]

df[df["user"]=="想太多de猫"].sample(10)

<table> 
 <thead> 
  <tr> 
   <th></th> 
   <th>episodes</th> 
   <th>user</th> 
   <th>comment</th> 
   <th>grade</th> 
   <th>date</th> 
   <th>likecounts</th> 
  </tr> 
 </thead> 
 <tbody> 
  <tr> 
   <th>17104</th> 
   <td>第1期</td> 
   <td>想太多de猫</td> 
   <td>爱吃辣的都是美女</td> 
   <td>v4</td> 
   <td>00:51:54</td> 
   <td>9</td> 
  </tr> 
  <tr> 
   <th>20881</th> 
   <td>第2期</td> 
   <td>想太多de猫</td> 
   <td>如果这都是咸鱼</td> 
   <td>v4</td> 
   <td>00:09:45</td> 
   <td>10</td> 
  </tr> 
  <tr> 
   <th>1489</th> 
   <td>第1期</td> 
   <td>想太多de猫</td> 
   <td>领带神马的太高端</td> 
   <td>v4</td> 
   <td>00:04:42</td> 
   <td>12</td> 
  </tr> 
  <tr> 
   <th>5408</th> 
   <td>第1期</td> 
   <td>想太多de猫</td> 
   <td>谦虚不等于诚实哦</td> 
   <td>v4</td> 
   <td>00:15:49</td> 
   <td>12</td> 
  </tr> 
  <tr> 
   <th>57919</th> 
   <td>面试篇</td> 
   <td>想太多de猫</td> 
   <td>大多数父母的想法</td> 
   <td>v4</td> 
   <td>00:14:54</td> 
   <td>7</td> 
  </tr> 
  <tr> 
   <th>60506</th> 
   <td>面试篇</td> 
   <td>想太多de猫</td> 
   <td>越看越自卑</td> 
   <td>v4</td> 
   <td>00:21:36</td> 
   <td>13</td> 
  </tr> 
  <tr> 
   <th>45714</th> 
   <td>第3期</td> 
   <td>想太多de猫</td> 
   <td>这是绝对的高质量对抗</td> 
   <td>v0</td> 
   <td>00:37:00</td> 
   <td>2</td> 
  </tr> 
  <tr> 
   <th>21397</th> 
   <td>第2期</td> 
   <td>想太多de猫</td> 
   <td>瞿泽林这句对人性的关怀</td> 
   <td>v4</td> 
   <td>00:11:05</td> 
   <td>49</td> 
  </tr> 
  <tr> 
   <th>59906</th> 
   <td>面试篇</td> 
   <td>想太多de猫</td> 
   <td>假发了解一下</td> 
   <td>v4</td> 
   <td>00:19:53</td> 
   <td>13</td> 
  </tr> 
  <tr> 
   <th>39564</th> 
   <td>第3期</td> 
   <td>想太多de猫</td> 
   <td>个人觉得直接问更好</td> 
   <td>v0</td> 
   <td>00:21:49</td> 
   <td>1</td> 
  </tr> 
 </tbody> 
</table>

### 3.会员等级分布 ###

df_g = df['grade'].value_counts().sort_values(ascending=True)
    df_g

v8       12
    v7      135
    v6     1496
    v5     2097
    v2     2499
    v4     2957
    v1     3220
    v3     3403
    v0    54286
    Name: grade, dtype: int64

x = df_g.index.tolist()
    y = df_g.values.tolist()
    df_g = get_pyechart(x=x,y=y,chart='pie',title='会员等级分布',size=14,pos='right',theme=ThemeType.DARK)
    df_g

![在这里插入图片描述][watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3FxXzQ0NzYwOTEy_size_16_color_FFFFFF_t_70 2]

### 词云图讨论 ###

import stylecloud
    import jieba
    import os 
    from IPython.display import Image # 用于在jupyter lab中显示本地图

# 定义分词函数
    def get_cut_words(content_series):
        # 读入停用词表
        stop_words = [] 
        with open(r"D:/Pandas/已学习/如何制作stylecloud词云？/stop_words.txt", 'r', encoding='utf-8') as f:
            lines = f.readlines()
            for line in lines:
                stop_words.append(line.strip())
        # 添加关键词
        my_words = ['撒老师', '范丞丞','第一季']  
        for i in my_words:
            jieba.add_word(i) 
        # 自定义停用词
        my_stop_words = ['好像', '真的','感觉']   
        stop_words.extend(my_stop_words)               
        # 分词
        word_num = jieba.lcut(content_series.str.cat(sep='。'), cut_all=False)
        # 条件筛选
        word_num_selected = [i for i in word_num if i not in stop_words and len(i)>=2]
        return word_num_selected
    # 绘制词云图
    text1 = get_cut_words(content_series=df['comment'])
    stylecloud.gen_stylecloud(text=' '.join(text1), max_words=100,
                              collocations=False,
                              font_path='字酷堂清楷体.ttf',
                              icon_name='fas fa-dog',
                              size=512,
                              #palette='matplotlib.Inferno_9',
                              output_name='offer.png')
    Image(filename='offer.png')

Building prefix dict from the default dictionary ...
    Loading model from cache C:\Users\ADMINI~1\AppData\Local\Temp\jieba.cache
    Loading model cost 2.044 seconds.
    Prefix dict has been built successfully.

![\[外链图片转存失败,源站可能有防盗链机制,建议将图片保存下来直接上传(img-PL3JJJhN-1608626592533)(demo-offer\_files/demo-offer\_39\_1.png)\]][img-PL3JJJhN-1608626592533_demo-offer_files_demo-offer_39_1.png]

### 4.人物被提及 ###

丁辉,詹秋怡,王骁,朱一暄,瞿泽林,李晋晔,王颖飞,刘煜成

#df.str.contains('')虽然不能用来判断，但是可以进行聚合操作
    df_talk = ['丁辉','詹秋怡','王骁','朱一暄','瞿泽林','李晋晔','王颖飞','刘煜成']
    dft = [df['comment'].str.contains(i).sum() for i in df_talk]
    df_t = pd.DataFrame(data={ 'people':df_talk,'count':dft})
    df_t = df_t.sort_values('count',ascending=False)
    x = df_t['people'].tolist()
    y = df_t['count'].tolist()
    df_t = get_pyechart(x=x,y=y,chart='bar',title='被提及次数',size=16,pos='top',theme=ThemeType.DARK)
    df_t

![在这里插入图片描述][watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3FxXzQ0NzYwOTEy_size_16_color_FFFFFF_t_70 3]

### 情感分析 ###

import paddlehub as hub
    #这里使用了百度开源的成熟NLP模型来预测情感倾向
    senta = hub.Module(name="senta_bilstm")
    texts = df['comment'].tolist()
    input_data = { 'text':texts}
    res = senta.sentiment_classify(data=input_data)
    df['pos'] = [x['positive_probs'] for x in res]
    #重采样至15分钟
    df.index = pd.to_datetime(df['date'])
    data = df.resample('15min').mean().reset_index()
    
    # #给数据表添加调色板
    # import seaborn as sns
    # color_map = sns.light_palette('orange', as_cmap=True) #light_palette调色板
    # data.style.background_gradient(color_map)

[2020-12-22 16:27:46,410] [    INFO] - Installing senta_bilstm module
    [2020-12-22 16:27:46,436] [    INFO] - Module senta_bilstm already installed in C:\Users\Administrator\.paddlehub\modules\senta_bilstm
    [2020-12-22 16:27:57,111] [    INFO] - Installing lac module
    [2020-12-22 16:27:57,579] [    INFO] - Module lac already installed in C:\Users\Administrator\.paddlehub\modules\lac

df[:10]

x = data["date"].to_list()
    y = list(data["pos"].round(2))
    df_p = get_pyechart(x=x,y=y,chart='line',title='情感倾向',size=16,pos='top',theme=ThemeType.DARK)
    df_p

![在这里插入图片描述][watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3FxXzQ0NzYwOTEy_size_16_color_FFFFFF_t_70 4]

# 数据集： #

链接：[https://pan.baidu.com/s/1p4O8-SF-IVqnJkH-5Fp3jw  
][https_pan.baidu.com_s_1p4O8-SF-IVqnJkH-5Fp3jw]提取码：love  
复制这段内容后打开百度网盘手机App，操作更方便哦

# 总结 #

情感分析会比较耗时间

[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3FxXzQ0NzYwOTEy_size_16_color_FFFFFF_t_70]: /images/20221120/27d630c5329f4a219a948116088935e3.png
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3FxXzQ0NzYwOTEy_size_16_color_FFFFFF_t_70 1]: /images/20221120/d4d5af3c08f841f1963c560b77350c41.png
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3FxXzQ0NzYwOTEy_size_16_color_FFFFFF_t_70 2]: https://img-blog.csdnimg.cn/20201222165917851.png?x-oss-process=image/watermark,type_ZmFuZ3poZW5naGVpdGk,shadow_10,text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3FxXzQ0NzYwOTEy,size_16,color_FFFFFF,t_70
[img-PL3JJJhN-1608626592533_demo-offer_files_demo-offer_39_1.png]: https://img-blog.csdnimg.cn/20201222165827726.png?x-oss-process=image/watermark,type_ZmFuZ3poZW5naGVpdGk,shadow_10,text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3FxXzQ0NzYwOTEy,size_16,color_FFFFFF,t_70
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3FxXzQ0NzYwOTEy_size_16_color_FFFFFF_t_70 3]: https://img-blog.csdnimg.cn/20201222165725284.png?x-oss-process=image/watermark,type_ZmFuZ3poZW5naGVpdGk,shadow_10,text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3FxXzQ0NzYwOTEy,size_16,color_FFFFFF,t_70
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3FxXzQ0NzYwOTEy_size_16_color_FFFFFF_t_70 4]: https://img-blog.csdnimg.cn/20201222170041469.png?x-oss-process=image/watermark,type_ZmFuZ3poZW5naGVpdGk,shadow_10,text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3FxXzQ0NzYwOTEy,size_16,color_FFFFFF,t_70
[https_pan.baidu.com_s_1p4O8-SF-IVqnJkH-5Fp3jw]: https://pan.baidu.com/s/1p4O8-SF-IVqnJkH-5Fp3jw