If you guys are getting a lot of spam, I found something that might interest you. I read in a technical journal (EDN) about a new spam filter (SpamBayes) that is mathematically based. It is a Bayesian (http://spambayes.sourceforge.net/glossary) anti-spam filter, initially based on the work of Paul Graham (http://www.paulgraham.com/spam.html).
When you first fire SpamBayes up, you show it folders that have spam in it and a folder that does not have spam. It generates a hypotheses based on the messages YOU consider spam.
Once done, the program checks each message as you get it. If it calculates the message is spam, it put it in the junk email folder. If it may be spam, it put it in a Maybe folder and you can check it out. If you feel this is spam when you look in the Maybe folder, you tell the program it is spam and it adjusts its hypotheses.
When I first activated it, it captured 100% of the spam I was getting (I had a LOT of spam samples saved up.) After about two weeks, I only had two emails in my Maybe folder. I average about 25 spam emails per day and this program is nailing every one.
Oh, the program is freeware (donations gladly accepted.)
http://spambayes.sourceforge.net
When you first fire SpamBayes up, you show it folders that have spam in it and a folder that does not have spam. It generates a hypotheses based on the messages YOU consider spam.
Once done, the program checks each message as you get it. If it calculates the message is spam, it put it in the junk email folder. If it may be spam, it put it in a Maybe folder and you can check it out. If you feel this is spam when you look in the Maybe folder, you tell the program it is spam and it adjusts its hypotheses.
When I first activated it, it captured 100% of the spam I was getting (I had a LOT of spam samples saved up.) After about two weeks, I only had two emails in my Maybe folder. I average about 25 spam emails per day and this program is nailing every one.
Oh, the program is freeware (donations gladly accepted.)
http://spambayes.sourceforge.net