SpamBayes

The SpamBayes project is working on developing a statistical (commonly, although a little inaccurately, referred to as Bayesian) anti-spam filter, initially based on the work of Paul Graham. The major difference between this and other, similar projects is the emphasis on testing newer approaches to scoring messages. While most anti-spam projects are still working with the original graham algorithm, we found that a number of alternate methods yielded a more useful response. This is documented on the background page.

I have been using this spam filter for my Outlook inbox and yes the application has been able to filter the junk 98% of the time. This works cause I have trained it to identify the bad from the good. Within some time has been effective in clearing most of the junk from my inbox. Its dynamic and scans the emails in realtime so that it gets filtered to the junk mail folder as soon as it reaches your inbox.
I will be trying out another open source anti spam software Spamato now which seems to be very promising. But if you dont have much time to wait for the next review please go ahead and check out SpamBayes. This is one open source application which is a must have in these times! Bye bye spam...

Dug up some reviews and tests on SpamBayes. Comes up as a winner!
1. Winning the War on spam: Comparison of Bayesian spam filters
2. What people are saying about (us) SpamBayes (from their own site)