【战狼2】影评分析

最近大家都在讨论【战狼2】，大都给了很高的评价。媒体也称这部电影为‘最接近好来屋大片的国产电影’。今天做了个简单的评价，结果显示战狼2的负面评价远超过正面评价。

所需R包

library(rvest)library(stringr)library(tidyverse)library(wordcloud2)library(jiebaR)

数据抓取

一共抓取了1000条豆瓣评论.

url0='https://movie.douban.com/subject/26363254/comments?'
page='status=P'
name=name_href=score=time=com_score=commen=NULL
ite=1
while(ite<=50){

  url=paste0(url0,page)

  web <- read_html(url)

  page=web%>%html_nodes('.next')%>%html_attr('href')

  tt1=web%>%html_nodes('span.comment-info>a')

  name=c(name,tt1%>%html_text())

  name_href=c(name_href,tt1%>%html_attr('href'))

  score=c(score,web%>%html_nodes('div.comment-item > div:nth-child(2) > 

   h3:nth-child(1) > span:nth-child(2) > span:nth-child(3)')%>%

   html_attr('title'))

  time=c(time,web%>%html_nodes('div.comment-item > div:nth-child(2) > 

   h3:nth-child(1) > span:nth-child(2) > span:nth-child(4)')%>%

   html_attr('title'))

  com_score=c(com_score,web%>%html_nodes('.comment-vote>span')%>%html_text())

  commen=c(commen,web%>%html_nodes('.comment>p')%>%html_text())

  ite=ite+1}

datt=data.frame(name,score,time,com_score,commen,name_href,stringsAsFactors = F)

names(datt)=c('用户','评价','时间','评论得分','评论','超链接')

write.csv(datt,'datt.csv')

DT::datatable(datt)

简单数据分析

战狼二远没有你想的好

library(ggplot2)

datt1=as.data.frame(table(datt$评价))

ggplot(datt1,aes(reorder(Var1, Freq),Freq))+

  geom_bar(stat="identity", fill="steelblue")+

  geom_text(aes(label= Freq), vjust=1, hjust=1.5, color="white", size=3.5)+

  labs(x='评价',y='总数')+

  coord_flip()+

  labs(title='【战狼2】用户评价条形图', subtitle=R.version.string, 

   caption='@EASYCHARTS')+theme_minimal()

差评者都说了什么？

datt2 = datt[datt$评价=='很差',]

comment1=datt2$评论%>%str_trim()

txt1=wolf[comment1]

datx1=as.data.frame(table(txt1),stringsAsFactors = F)[-1:-7,]

datx1[str_length(datx1$txt)>1,]

txda0=arrange(datx1,desc(Freq))

txda10=txda0[str_length(txda0$txt)>1,]

wordcloud2(datx1,color="random-light",backgroundColor = 'black')

总体评价

#stop_words.txt是把jiebaRD底层的stop_words.utf-8篡改wolf<-worker('mix',

  stop_word="D:/Program Files/R/R-3.3.2/library/jiebaRD/dict/stop_words.txt")

comment=datt$评论%>%str_trim()

comment[1]

txt=wolf[comment]

datx=as.data.frame(table(txt),stringsAsFactors = F)[-1:-7,]

datx[str_length(datx$txt)>1,]

txda=arrange(datx,desc(Freq))

txda1=txda[str_length(txda$txt)>1,]



wordcloud2(txda1,color="random-light",backgroundColor = 'black')

注意：先运行总体评价的代码，在运行差评的部分.