Today I decided to revisit the Bach Choral Harmony data set and look at chord progression transitions.
In order to do this I wrote a small program that tallies the movement from one chord to another, and then outputs a Graphviz dot file that can be turned into an image.
Today being a Saturday, I decided to noodle around with the GD::Simple module and constructed rudimentary Chernoff faces - Cartoon data visualization.
In 1968, Aristid Lindenmayer introduced "L-systems" to the world. Basically, this is an iterative rewriting system of rules that operates on a string beginning with an "axiom" or initial string.
I decided to write my own sentiment analysis library called Lingua::EN::Opinion and use it to investigate the KJV Bible Book of Revelation.
How do the books of the KJV Bible compare to each other word for word? Are Old Testament books more similar to other Old Testament books? Same for the New Testament? What about the Greek (Nestle-Aland 26th/27th edition) translation of the New Testament?
Naturally, I wrote a little perl program of the classic NLP measure to tell me: cosine-similarity. It is not the most efficient but it works perfectly!