搜索结果词频可视化: R语言wordcloud（词云图）

 先上成品图，观摩一下。

对一段文字做词频统计，然后根据词频做成词云图，途中文字大小与词频成正比。

结果实际上是一个交互html文件，也可输出为图形。

 上代码

library(wordcloud2)
library(dplyr)

filePath = "COVID1.txt"    #搜索结果存为txt
text = readLines(filePath)
txt = text[text!=""]       #校验是否为空文件

txt<-iconv(txt,"WINDOWS-1252","UTF-8")  #转码
txt = tolower(txt)
txtList = lapply(txt, strsplit," ")
txtChar = unlist(txtList)

#清洗数据
txtChar = gsub("\\.|,|\\!|:|;|\\?|(|)","",txtChar) #去除常见符号（.,!:;?)
#txtChar<-gsub("[^a-zA-Z]"," ",txtChar)
txtChar = txtChar[txtChar!=""]

data = as.data.frame(table(txtChar))
data$txtChar<-as.character(data$txtChar)
data<-data[-which(nchar(data[,1])<=2),]            #过滤长度<=2的单词
colnames(data) = c("Word","freq")
ordFreq = data[order(data$freq,decreasing=T),]     #词频排序

#过滤常见词
filePath = "filter200.csv"                          #常用虚词列表              
df = read.csv(filePath,header = T)
Word = select(df,Word)
antiWord = data.frame(Word,stringsAsFactors=F)
result = anti_join(ordFreq,antiWord,by="Word") %>% arrange(desc(freq)) #取差集
head(result)
wordcloud2(data = result)

 想要 常用过滤词文件 filtere200.csv 的同学，请回复，楼主发免费下载链接。