I don't know if the Mozilla one comes with a pre-filled database, but this technique (filtering on the words on existing spam to spot more spam) requires a corpus of spam, so start saving it in a folder rather than binning it so that when you have a Bayesian filter you have something to work with.
Incidentally the bloke with the patent on this algorithm was in the same class as Damian and me at college.
Don't enable a preview pane when dragging it: most contain a zero pixel graphic with your e-mail address in a query string that act as a read-receipt to the spammer when they check their server logs.
![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png)
chiming
no subject
Date: January 2nd, 2003 03:30 am (UTC)From:Incidentally the bloke with the patent on this algorithm was in the same class as Damian and me at college.
Don't enable a preview pane when dragging it: most contain a zero pixel graphic with your e-mail address in a query string that act as a read-receipt to the spammer when they check their server logs.