Think about it. I get a lot of spam for purchasing insurance. If I just filtered based on the word insurance, I would stop a lot of the spam that I currently receive. BUT, it would also filter out the e-mails I receive from Progressive insurance reminding me to pay my car insurance. So, instead of doing that, this program looks at the entire e-mail, and compares it to other e-mails that I have received, coming up with bayesian probabilities for words in each particular group.

Progressive's e-mails have my name written in them. So do a lot of other e-mails that I receive, but no spam ever has my name in it. That means that it is highly probable that an e-mail with my name in it is not spam. If you want to know how the math works, you can probably ask Gordon, considering he's the math professor.

I'm currently working on setting up this program so it can run as a server on my Linux box. This is also a good learning task for me, as I'm going to have to modify the program a little bit to do that.

In weekend recaps, I listened to The Campbell Brothers on Saturday night. They sure can jam, but me and a bunch of the people working the concert realized that white people cannot keep a beat. Oh well. The music was good, the cookies even better.

That's just about all I did this weekend.