library(httr)
library(jiebaRD)
library(jiebaR)
library(tm)
library(dplyr)
library(wordcloud)

文本清理

將欲分析之本機txt檔讀入，使用 tm package進行資料的清理，去除無效的用詞如：特殊符號、停留詞、英文字、數字等，使用 jiebaR package進行斷詞

#load files
filenames = list.files(getwd(),pattern = "*.txt")
files = lapply(filenames,readLines)
docs = Corpus(VectorSource(files))

#remove meaningless character
toSpace = content_transformer(function(x,pattern) {
  return (gsub(pattern," ",x))
})

docs = tm_map(docs,toSpace,"【")
docs = tm_map(docs,toSpace,"】")
docs = tm_map(docs,toSpace,"\\[")
docs = tm_map(docs,toSpace,"\\]")
docs = tm_map(docs,toSpace,"的")
docs = tm_map(docs,toSpace,"了")
docs = tm_map(docs,toSpace,"有")
docs = tm_map(docs,toSpace,"可")
docs = tm_map(docs,toSpace,"或")
docs = tm_map(docs,toSpace,"月")
docs = tm_map(docs, removePunctuation)
docs = tm_map(docs,removeNumbers)

#words cut
cutter = worker()
new_user_word(cutter,"捷運站")

tokenizer = function(x) {
  segment(x,cutter)
}

freqFrame = as.data.frame(table(unlist(lapply(docs,tokenizer)))) %>%
  arrange(desc(Freq))

將文本處理結果以文字雲的方式視覺化

#build a wordcloud
par(family=("Heiti TC Light"))
wordcloud(freqFrame$Var1,freqFrame$Freq,
          scale=c(5,0.1),min.freq=50,max.words=150,
          random.order=TRUE, random.color=FALSE,
          rot.per=.1, colors=brewer.pal(8, "Dark2"),
          ordered.colors=FALSE,use.r.layout=FALSE,
          fixed.asp=TRUE)