帝吧facebook战报（wordcloud）

API

library(devtools)
install_github("Rfacebook", "pablobarbera", subdir="Rfacebook")
require("Rfacebook")
fb_oauth <- fbOAuth(app_id="xx", app_secret="xx"
                    ,extended_permissions = TRUE)

DATA

• 以三立新闻facebook主页最近（2016/01/22）100条post下的comment数据为样本.
  
• 三立新闻的facebook id为110699089014688.

sanli<-getPage(110699089014688,token = fb_oauth,n=100)
post.id <- sanli$id

commentExtract<-function(x){
  getPost(post=x,token = fb_oauth)$comments$message
}

comments <- lapply(post.id,FUN = commentExtract)#list[1:100]

data clean

library(Rwordseg,quietly = TRUE)
comments<-unlist(comments)
cnsegment<-function(x){
  tryCatch(segmentCN(x),
           error = function(e) return(""))
}
segment.cn<- unlist(lapply(comments, cnsegment))
word = lapply(X = segment.cn, FUN = strsplit, " ")

v = table(unlist(word))
v = sort(v, decreasing = T)
v[1:100]
head(v)
d = data.frame(word = names(v), freq = v)

visualization

par(family = "STHeiti")
require(wordcloud)
dd = d
op = par(bg = "lightyellow")
# grayLevels = gray((dd$freq)/(max(dd$freq) + 140))
# wordcloud(dd$word, dd$freq, colors = grayLevels)
colors=brewer.pal(8,"Dark2")
wordcloud(dd$word, dd$freq, max.words=100,
          scale=c(10,3),min.freq=-Inf,
          colors=colors,random.order=F,random.color=F,ordered.colors=F
          )
par(op)