InformationOverlord

Sometimes keeping up with all of the reading I’d like to do leaves me crushed by information overload, so for Programming A to Z this week I built a simple program to tell me whether or not I should bother reading a given document. Our assignment was to augment, modify, or replace one of our in-class examples, so I modified Adam’s BayesClassifier program. My version of the classifier, InformationOverlord, requires input of at least two texts; it compares the specified text to the first (one or more) texts, and advises you to read or discard the specified text based on its comparison score. For instance, if I particularly enjoyed Mark Twain’s Roughing It and was wondering if I might also like The Wasteland by T.S. Eliot I could use InformationOverlord to make a recommendation for me:

overlord_drivel1

Don't bother reading this drivel.

On the other hand, if I ask it to tell me whether or not I ought to read the works of Jane Austen, based on my interest in Shakespeare’s sonnets, it tells me something different:

You should definitely read this.

You should definitely read this.

Reply