CODEWITHSUNDEEP
rsentiment-analysisstemming
user3456230
user3456230

Reputation: 217

error message while stemming for sentiment analysis

I do stemming on my dataset for sentiment analysis and I got this error message

"Error in structure(if (length(n)) n else NA, names = x) : 'names' attribute [2] must be the same length as the vector [1]"

Please help!

myCorpus<-Corpus(VectorSource(Datasetlow_cost_airline$text))
# Convert to lower case
myCorpus<-tm_map(myCorpus,tolower)
# Remove puntuation
myCorpus<-tm_map(myCorpus,removePunctuation)
# Remove numbers
myCorpus<-tm_map(myCorpus,removeNumbers)
# Remove URLs ?regex = regular expression ?gsub = pattern matching
removeURL<-function(x)gsub("http[[:alnum:]]*","",x)
myCorpus<-tm_map(myCorpus,removeURL)
stopwords("english")
# Add two extra stop words: 'available' and 'via'
myStopwords<-c(stopwords("english"),"available","via","can")
# Remove stopwords from corpus
myCorpus<-tm_map(myCorpus,removeWords,myStopwords)
# Keep a copy of corpus to use later as a dictionary for stem completion
myCorpusCopy<-myCorpus
# Stem word (change all the words to its root word)
myCorpus<-tm_map(myCorpus,stemDocument)
# Inspect documents (tweets) numbered 11 to 15
for(i in 11:15){
cat(paste("[[",i,"]]",sep=""))
writeLines(strwrap(myCorpus[[i]],width=73))
}
# Stem completion
myCorpus<-tm_map(myCorpus,stemCompletion,dictionary=myCorpusCopy)

Upvotes: 0

Views: 780

Answers (1)

Ben
Ben

Reputation: 42333

There seems to be something odd about the stemCompletion function in tm version 0.6. There is a nice workaround here that I've used for this answer. In brief, replace your

# Stem completion
myCorpus <- tm_map(myCorpus, stemCompletion, dictionary = myCorpusCopy) # use spaces!

with

# Stem completion
stemCompletion_mod <- function(x,dict) {
  PlainTextDocument(stripWhitespace(paste(stemCompletion(unlist(strsplit(as.character(x)," ")), dictionary = dict, type = "shortest"), sep = "", collapse = " ")))
}
# apply workaround function 
myCorpus <- lapply(corpus, stemCompletion_mod, myCorpusCopy)

If that doesn't help then you'll need to give more details and a sample of your actual data.

Upvotes: 1

Related Questions