R语言 迪士尼点评文本挖掘

setwd("D:\\迪士尼点评文本挖掘")

#getwd()可查看目前的工作路径

#加载包()

library(DBI)

#library(RMySQL)

library(rJava)

library(openxlsx)

library(stringr)

library(xlsxjars)

library(reshape)

library(readxl)

library(xlsx)

library(sqldf)

library(wordcloud)

library(Rwordseg) #加载分词包

library(tm)

library(tmcn)

library(jiebaRD)

library(jiebaR) #里面的segmentC用于分词

#library(wordcloud2)

###########################读入表格###############################

comment_01 <- read_excel("点评文本挖掘.xlsx",1) #读入原始文本

comment_01 <- comment_01[,'点评内容'] #仅保留文本字段

stopwords_01 <- read_excel("停用词汇总.xlsx",1) # 读入停用词文本

#str函数，即structure,紧凑的显示对象内部结构，即对象里有什么。作用跟head相似

#str(comment_01)

#添加搜狗词汇词典 查看已安装的词典用 ：listDict()。卸载词典：uninstallDict()。

installDict(dictpath = '旅游词汇大全【官方推荐】.scel',dictname = 'Vocabulary_books', dicttype = 'scel') #旅游词汇大全

installDict(dictpath = 'disney.scel',dictname = 'disney', dicttype = 'scel') #迪士尼词汇大全

installDict(dictpath = '自定义词典.txt',dictname = 'dictionary_01') #迪士尼词汇大全

#uninstallDict("disney")

#uninstallDict("Vocabulary_books")

#uninstallDict("dictionary_01")

#加词

#uninstallDict(disney)

listDict()

insertWords(c("排队","不满意","非常满意","很好","不方便","非常好","很棒","驴妈妈","不舒服","不值","七个小矮人","飞越地平线","创极速光轮","米奇大街","奇想花园","梦幻世界","探险岛","宝藏湾","明日世界","巴斯光年星际营救","喷气背包飞行器","太空幸会史迪奇","星球大战远征基地","皮克斯玩具总动员","快速通道","加勒比海盗","灯光秀","飞跃地平线"))

#去掉字母和数字

comment_02 <- gsub('[0-9a-zA-Z]','',comment_01)

#分词

segword <- unlist(lapply(X=comment_02,FUN=segmentCN))

#创建停止词

#head(stopwords_01) #查看数据

#class(stopwords_01) #查看变量类型，可知是属于数据框类型

#segword[1:10]

#需要将数据框格式的数据转化为向量格式

stopwords_01<- as.matrix(stopwords_01[,1])

stopwords_01<- as.vector(stopwords_01[,1])

#自定义删除停止词的函数

removeStopWords <- function(x,stopwords) {

temp <- character(0)

index <- 1

xLen <- length(x)

while (index <= xLen) {

if (length(stopwords[stopwords==x[index]]) <1)

temp<- c(temp,x[index])

index <- index +1

}

temp

}

#删词

segword3 <-lapply(segword,removeStopWords,stopwords_01)

#绘制文字图

word_freq <- createWordFreq(unlist(segword3)) #createWordFreq函数来自于tmcn包

opar <-par(no.readonly = TRUE)

par(bg = 'black')

#绘制出现频率最高的前100个词

wordcloud(words=word_freq$word,freq=word_freq$freq,max.words=100,random.color=TRUE,colors=rainbow(n=7))

par(opar)

##############################数据输出##############################

result_filename<-paste(Sys.Date(),'数据',".xlsx",sep="")

write.xlsx(word_freq,result_filename,sheetName='明细')