分析10万条弹幕，告诉你古董局中局2这部鉴宝题材剧咋样？

先导入数据处置方式和相关的作图库小说题材。

#导入库

importnumpiasnp

importpandaaspd

importmatplotlib.pyplotasplt

importjieba

importjieba.analyse

fromsnownlpimportSnowNLP

frompyecharts.chartimportBar,Pie,Line,WordCloud,Page

frompyechartimportopt影视题材ionasopts

01弹视频题材幕基本概况

爬取到弹幕是这样的的子的一共包罗了以下有关相关信息：集数、议论id视频题材用户名、vip品阶、议论其他其他内容、议论时间点和评论点赞。

#读入数据

df_1=pd.read_c瀑布题材sv\’./data/古董局中局2腾讯弹幕.csv\’,encoding=\’utf-8瀑布题材\’,engine=\’python\’

df_1.head

一共有107570条弹幕数据。

df_1.info

<class\’pandas.core.fram短片题材e.DataFrame\’>

RangeIndex:107570entries,0to107569

Datacolumntotal7column:

episod107570non-nulint64

comment_id107570non-nulint64

oper_nam25539non-nulobject

vip_degre107570non-nulint64

cont107570non-nulobject

time_point107570non-nulint64

up_count107570non-nulint64

dtypes:int645,object2

memoriusage:5.7+MB

02弹幕字数分布

#寻思字数

word_num=df_1[\’content\’].applilambdax:lenx

#分箱

bin=[0,5,10,15,20,25,30,35,40,45,50,549]

word_num_cut=pd.cutword_num,bin.value_count.sort_index

#柱形图

bar1=Barinit_opts=opts.InitOptwidth=\’1350px\’,height=\’750px\’

bar1.add_xaxiword_num_cut.index.astyp\’str\’.tolist

bar1.add_yaxi””,word_num_cut.values.tolist,category_gap=\’4%\’

bar1.set_global_opttitle_opts=opts.TitleOpttitle=”弹幕发送字数分布”,

visualmap_opts=opts.VisualMapOptmax_=38107

bar1.rend

03弹幕情感分析

defnlp_scorex:

“””

效用短片题材：可可获取心情评分

“””

sn=SnowNLPx

returnsn.sentiments

#寻思心情得分

df_1[\’nlp_score\’]=df_1[\’content\’].maplambdax:nlp_scorex

#分箱

score_bin=[0,0.1,0.2,小说题材0.3,0.4,0.5,0.6,0.7,0.8,0.9,1]

score_cut=pd.cutdf_1[\’nlp_score\’],bins=score_bin

score_cut=score_cut.value_count.sort_index

#绘制折线图

line1=Lineinit_opts=opts.InitOptwidth=\’1350px\’,height=\’750px\’

line1.add_xaxiscore_cut.index.astyp\’str\’.tolist

line1.add_yaxi\’\’,score_cut.values.tolist,

areastyle_opts=opts.A reaStyleOptopacity=0.5,

label_opts=opts.LabelOptis_show=Fals

line1.set_global_opttitle_opts=opts.TitleOpttitle=\’弹幕心情评分分布[0~1]\’,

#toolbox_opts=opts.ToolboxOpt,

visualmap_opts=opts.VisualMapOptmax_=22288

line1.set_series_optlinestyle_opts=opts.LineStyleOptwidth=4

line1.rend

04弹幕关注更多更多人数数量占比

defcalculate_numwords_list:

“””

效用：给定关键词列表，寻思弹幕出现次数

“””

pattern=\’|\’.joinwords_list

num=intdf_1[\’content\’].str.containpattern.sum

returnnum

#关键词

xiayu=[\’许下愿望\’,\’夏雨\’,\’下雨\’]

weichen=[\’药不然\’,\’魏晨\’,\’药二爷\’,\’晨晨\’,\’老魏\’,\’晨哥\’,\’晨\’]

laochaofeng=[\’老朝奉\’,\’老朝凤\’,\’老朝\’,\’老嘲讽\’,\’老巢\’]

A liya=[\’黄烟烟\’,\’阿丽亚\’,\’烟烟\’]

liangj=[\’梁静\’,\’素姐\’]

liuyim=[\’一鸣\’,\’毕彦君\’]

all_list=[xiayu,weichen,laochaofeng,A liya,liangjing,liuyiming]

danmu_nam=[\’夏雨\’,\’魏晨\’,\’老朝奉\’,\’阿丽亚\’,\’梁静\’,\’刘一鸣\’]

danmu_num=[calculate_numiforiinall_list]

#柱形图

bar2=Barinit_opts=opts.InitOptwidth=\’1350px\’,height=\’750px\’

bar2.add_xaxidanmu_nam

bar2.add_yaxi\’数量\’,danmu_num

bar2.set_global_opttitle_opts=opts.TitleOpt\’弹幕主要关注更多更多人物散布\’,

visualmap_opts=opts.VisualMapOptmax_=20000

bar2.rend

结语

现在这部《古2还在更新中，人各也也可以等都更新完了看，终究免费看剧岂不不香吗？同时C君在这里还是很举荐人各去玩下这个《古董局中局》桌游，确比较烧脑的

本文出品：CDA 数据分析师（ID:cdacdacda

原文链接：https://blog.csdn.net/yoggieCDA /article/details/106228954?ops_request_misc=%257B%作文题材2522request%255Fid%2522%253A %2522166856496116782425691787%2522%252C%2522scm%2522%253A %252220140713.130102334.pc%255Fblog.%2522%257D&request_id=166856496116782425691787&biz_id=0&utm_medium=distribute.pc_search_result.none-task-blog-2~blog~first_rank_ecpm_v1~times_rank-15-106228954-null-null.nonecase&utm_term=%E9%A 2%9影视题材8%E6%9D%90

未经允许不得转载：题材网 » 分析10万条弹幕，告诉你古董局中局2这部鉴宝题材剧咋样？