Write your own Bayesian classifier (in whatever language you want)

Let's talk about Bayesian classifiers. They're everywhere. They filter your spam, recommend new links, and categorize web pages.

Why are they so omnipresent? Because the math behind them is simple to understand, and simple to implement. So simple, in fact, that we'll do it in an hour.

I'll go over Bayes' theorem and how it relates to classification. I'll walk you through creating basic tokenization, training, and testing functions. Then we'll categorize stuff.

My implementation language will be Perl, but you're welcome to use whatever language you'd like.