bayes_motel
A multi-variate Bayesian classification engine in Ruby. Most popularly known for spam detection, Bayesian classification uses statistical methods to determine the proper category for a given document.
A multi-variate Bayesian classification engine in Ruby. Most popularly known for spam detection, Bayesian classification uses statistical methods to determine the proper category for a given document.
Comments
Curious as to the advantages/disadvantages or different design decisions compared to classifier (http://classifier.rubyforge.org/)?
Classifier is rather dumb in that it just takes a single String of text as input. BayesMotel will take multiple variables in a hash. This is useful if your document is more than just a body of text. For instance, a tweet is a hash of 20-30 different variables. Statistically, each variable can affect the document’s classification differently.
Sexy stuff!!
Post a comment