Word-clouds with R

What are the most used words in Sherlock Holmes? Let's make word-clouds and find out!

So I must admit that at first I tried this with the perl module Image::WordCloud. And it does ...ok, but not enough to be worthy of a journal post.

Enter the excellent language R!

I installed it with my operating system package manager. To make word-clouds, we need a couple additional libraries: "tm" (text mining) and "wordcloud" of course:

install.packages('tm')
install.packages('wordcloud')

To load them up, ready for use, do this:

library('tm')
library('wordcloud')

Read-in our text files as a corpus of documents:

docs <- Corpus(DirSource('~/Documents/lit/Sherlock-Holmes'))

Sanitize the text:

docs <- tm_map(docs, removeNumbers)
docs <- tm_map(docs, removePunctuation)
docs <- tm_map(docs, content_transformer(tolower))
docs <- tm_map(docs, removeWords, stopwords('english'))

Set the color palette to use:

pal <- brewer.pal(9, 'YlGnBu')
pal <- pal[-(1:4)]

Set the randomizer:

set.seed(123)

Set the filename to use:

png(file = 'Sherlock-wordcloud.png')

Create the cloud:

wordcloud(
    words = docs,
    scale = c(5, 0.1),
    max.words = 100,
    random.order = FALSE,
    rot.per = 0.35,
    use.r.layout = FALSE,
    colors = pal
)

Close the graphics device:

dev.off()

And what is the result?

How about we just look at one book, "The Hound of the Baskervilles":