If you used the analyzeSentiment function from SentimentAnalysis in R:

cor <- VCorpus(VectorSource(c('testing sentiment', 'this is great'))
sent <- analyzeSentiment(cor)

The names of the columns are:

"WordCount", "SentimentGI", "NegativityGI", "PositivityGI", "SentimentHE", "NegativityHE", "PositivityHE", "SentimentLM", "NegativityLM", "PositivityLM", "RatioUncertaintyLM" "SentimentQDAP", "NegativityQDAP", "PositivityQDAP"

Other than wordcount, what do all these labels mean? I didn't see any definitions in the vignette or docs.

  • $\begingroup$ Update: looks like the different GI, HE etc are lookup dictionaries, still would be good to know which they are and have links to sources. It would be nice to know what negativity and positivity are in relation to sentiment $\endgroup$ Dec 18, 2017 at 18:28

1 Answer 1


The different dictionaries are in the documentation for the function:

For example DictionaryHE is the "Dictionary with opinionated words from Henry’s Financial dictionary" (see DictionaryHE).


sentiment <- analyzeSentiment(document)   # Document is a Corpus
sentimentData <- sentiment$SentimentHE   # Generates analysis based on Henry's dictionary
plotSentiment(sentimentData) # Quick line plot of sentiment

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.