On Sun, 15 Feb 2004, Matthias F. Brandstetter uttered the following...
---------- quoting Grendel ----------
... trust me that bmf and bogofilter are better than spamassasin. I tested that with the last 1000 emails I recevied 200 were spam, and only 1 got through.
I have similar results using SpamAssassin over a year now. So why do you think is bmf better than SpamAssassin?
Well spam assasin is not very accurate, ie my ISP runs spam assasin but out of the > 4000 mails I get per day I find that about 80 spam get through, which is annoying, so I used bmf and it is very accurate, now only one spam per day gets through.
spammers are trying to cope up with this by adding a lot of dictionary worrds to the end of a spam but the filters learn about this.
If you have shell access at your ISP (with about 6MB free space) you could turn on Bayes Learning with Spam Assassin. That might require a little work on your part to train it and perhaps some scripts to automate some useful processes, but that is what I use and my false positive rate is 0% and my false negative rate is about 3% (meaning it catches 97% of the spam).
Tom Veldhouse
signature.asc
Description: OpenPGP digital signature
