Example: Gettysburg

The code and output below demonstrates some of RinRuby’s features. Ruby code counts the number of occurences of each word in Lincoln’s Gettysburg Address and filters out those occurring less than three times or shorter than four letters. R code — through the RinRuby library — produces a bar plot of the most frequent words and computes the correlation between the length of a word and the usage frequency. Finally, the computed correlation is printed by Ruby.  The text for the "gettysburg.txt" file is here.


   tally = Hash.new(0)
   File.open('gettysburg.txt').each_line do |line|
      line.downcase.split(/\W+/).each { |w| tally[w] += 1 }
   total = tally.values.inject { |sum,count| sum + count }
   tally.delete_if { |key,count| count < 3 || key.length < 4 }
   require "rinruby"
   R.keys, R.counts = tally.keys, tally.values
   R.eval <<EOF
      names(counts) <- keys
      barplot(rev(sort(counts)),main="Frequency of Non-Trivial Words",las=2)
      mtext("Among the #{total} words in the Gettysburg Address",3,0.45)
      rho <- round(cor(nchar(keys),counts),4)

   puts "The correlation between word length and frequency is #{R.rho}."


   The correlation between word length and frequency is -0.2779.

Subpages (1): gettysburg.txt