On Sun, 15 Feb 2004 20:09:58 +0000 Stroller <[EMAIL PROTECTED]> wrote:
> "only caught 92% of spam, with 1.16% false positives". So far I've been running it (bmf) for a week (yes wow, lol, but hear me out ;-) ), and it's gotten 100% of the approx 70+ spams I've received. Initially 2 "good" mails got through, but it's simply a matter of reprocessing those 2 incorrect mails through bmf again, stating they are incorrectly detected as spam, and you're set. It will take special precautions next time a similar one comes through. Incidentally both those e-mails were in Dutch, and the databases I had initially fed it were all in English mostly from mailing lists, so I'm not surprised ;-) For my "good" mail I fed it with most of my my friends / family / personal mails, and from all the mailing lists I belong to from this 2 weeks +-(can be downloaded from their monthly archives if need be). Ironically I didn't merge 1 single Dutch mail which caused an incorrect calculation, marking the mail as spam. I fed it with approx 500 spams I have received in the past couple of years from my"collection", and downloaded the latest 4 days archives of spam from ftp://spamarchive.org/pub/archives/submit/. I would definitely recommend this filter. I needed one that works easy, and spamassasin wasn't getting those spams with 1 link and 500 random words which counts at the moment of most of my spam, and bmf gets it every time! For each mail caught as spam, the database automatically updates itself with any new contents of that mail, making it learn as it's catching mails. If it makes a mistake (ie: you process it back through the filter once stating it's NOT spam) bmf corrects itself and stores the info that's required for it to pay spacial attention and kind of mail, and think 2/3/4 times before marking it as spam. It's a brilliant system I think. Anyways, just sharing my experience so far. Hope it was worth something Greetings Ralph -- http://axljab.homelinux.org/
pgp00000.pgp
Description: PGP signature
