Matador

From Miguel Jimenez:

Outlook 2003 comes with an integrated junk-mail filter that helps to prevent spam and other scums. I've spread my email addresses too much over newsgroups, forums, blogs, etc, and I use to receive between 200 and 300 emails a day. Most of them are spam. Almost of all them are not stopped by Outlook 2003 junk mail filter.

For this reason I started to investigate about filtering mail to detect spam. I saw an article by Paul Graham called "A Plan for Spam" that talked in early 2000 about Bayesian algorithms for the detection of spam. Bayesian algorithms are based on assigning a probability from the existing knowledge to any new item that comes to the filter. I've created my own implementation of this kind of algorithms, and optimized lots of things to improve the spam detection. Finally, created an Outlook 2003 Add In and plugged everything in to test it with my incoming mail. Results where amazing after my filter was trained. Almost 99.5% of my spam is stopped now.