library(rJava)
library(Rwordseg)
library(RColorBrewer)
library(wordcloud)
library(wordcloud2)
library(NLP)
library(tm)
installDict("D:/R/IT词汇大全.scel","computer",dicttype = "scel")
installDict("D:/R/IT计算机.scel","computer",dicttype = "scel")
installDict("D:/R/IT网管.scel","computer",dicttype = "scel")
installDict("D:/R/IT信息技术部.txt",dictname="IT信息技术部",dicttype = "text",load="ture")
lecture<-read.csv("D:/R/testword.txt", stringsAsFactors=FALSE,header=FALSE)
lecture1<-read.csv("D:/R/testgd.csv", stringsAsFactors=FALSE,header=FALSE)
words=unlist(lapply(X=lecture1, FUN=segmentCN))
word=lapply(X=words, FUN=strsplit, " ")
v=table(unlist(word))
v=rev(sort(v))
d=data.frame(词频=v)
#d=data.frame(词汇=names(v), 词频=v) #这样是三列
#dt1 <- subset(d,nchar(as.character(d$词频.Var1)) >1 & d$词频.Freq >=2)
dt <- subset(d,as.character(d$词频.Var1) !="的" & d$词频.Freq >=1)
write.csv(dt, file="D:/R/gongdan.csv", row.names=FALSE)
#wordcloud(d$词汇,d$词频.Freq,random.order=FALSE,random.color=TRUE,colors=mycolors,family="myFont")
jinb = system.file("examples/test.png",package = "wordcloud2")
wordcloud2(dt, figPath = jinb,size=0.5,color = 'random-light', minRotation = -pi/3, maxRotation = pi/3,rotateRatio = 0.8,fontFamily = "微软雅黑")
letterCloud(dt, word = "IT",wordSize = 1,backgroundColor = "white",color = 'lightgrey',minRotation = -pi/8, maxRotation = pi/2,rotateRatio = 0.8,fontFamily = "微软雅黑")
corpus = Corpus(VectorSource(words))
sample.dtm <- DocumentTermMatrix(corpus,control = list(wordLengths = c(1,Inf)))
df <- as.matrix(sample.dtm) #建立矩阵
df[df!=0]<-1
t_sample.dtm<- t(sample.dtm) #转置
sample.dtm2 <- removeSparseTerms(t_sample.dtm,sparse = 0.95)#去除稀疏词条，如果某一词出现在文档的概率少于（1-0.95）则不再保留到词频矩阵
