Created
July 10, 2017 10:48
-
-
Save yabyzq/6d27a5da39b0e7b9ca8df26604174704 to your computer and use it in GitHub Desktop.
TFIDF examples
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
library(RODBC) | |
library(tm) | |
library(wordcloud) | |
library(ggplot2) | |
library(ROracle) | |
a <- read.csv(file = 'C:/Users/eye1/Desktop/text.csv') | |
names(a) <- 'feedback' | |
narrative <- a$feedback | |
str(narrative) | |
head(narrative) | |
corpus <- Corpus(VectorSource(narrative))##assoc analysis, for speed, change this to narrative | |
corpus <- tm_map(corpus, content_transformer(tolower)) | |
corpus <- tm_map(corpus, removePunctuation) | |
corpus <- tm_map(corpus, stripWhitespace) | |
corpus <- tm_map(corpus, removeNumbers) | |
corpus <- tm_map(corpus, removeWords, c(stopwords("english"),"australia", "aust", "limited")) | |
dictCorpus <- corpus | |
corpus = tm_map(corpus, stemDocument); | |
#corpus = tm_map(corpus, stemCompletion, dictionary=dictCorpus); | |
#1 Simple Word Count | |
dtmSimple<-DocumentTermMatrix(corpus, control=list(wordLengths=c(4, 20)))#bounds=list(global=c(10, Inf))##c(floor(length(corpus)*0.05))) | |
dtmSMatrix <- as.matrix(dtmSimple) | |
freqSimple <- colSums(dtmSMatrix) | |
freqSimple <- sort(freqSimple, decreasing=TRUE) | |
head(freqSimple,15) | |
tail(freqSimple) ##findFreqTerms(dtm4,lowfreq=50) | |
wordcloud(names(freqSimple)[1:20], freqSimple[1:20],colors=brewer.pal(6,"Dark2")) | |
findFreqTerms(dtmSimple,10) | |
findAssocs(dtmSimple,"calls",0.1) | |
#2 Using tfidf | |
dtmTfidf<-DocumentTermMatrix(corpus,control = list(wordLengths=c(4, 20),#bounds=list(global=c(10, Inf)), | |
weighting = function(x) weightTfIdf(x, normalize = TRUE))) | |
dtmTMatrix <- as.matrix(dtmTfidf) | |
#Full Document Frequency | |
freqTfidf <- colSums(dtmTMatrix) | |
freqTfidf <- sort(freqTfidf, decreasing=TRUE) | |
head(freqTfidf,15) | |
tail(freqTfidf) ##findFreqTerms(dtm4,lowfreq=50) | |
#dissimilarity(dtmObject, method = "cosine") | |
#Top Words | |
wordcloud(names(freqTfidf)[1:15], freqTfidf[1:15],colors=brewer.pal(6,"Dark2")) | |
wf=data.frame(term=names(freqTfidf[1:15]),occurrences=freqTfidf[1:15]) | |
ggplot(wf, aes(reorder(term, -occurrences), occurrences)) + | |
geom_bar(stat="identity", fill='lightblue') + | |
theme(text = element_text(size=15, colour='grey100'), axis.text.x=element_text(angle=45, hjust=1),panel.background = element_rect(fill = "white")) |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment