On Sat, 10 Jan 2004 12:03:07 -0500 Bryan Phinney <[EMAIL PROTECTED]> wrote:
> On Saturday 10 January 2004 11:23 am, Richard Urwin wrote: > > > > I have to question that. I know it's better known, but I get far > > > better rates under POPFile than I did when I tried SA. POPFile > > > functions typically at 99.87% accuracy. I don't think you can better > > > that. > > > > That's a very impressive number, but I don't see what POPFile does that SA > > doesn't. You do need to train SA, and it doesn't start using Bayesian > > analysis until it's database has 200 messages in. I don't have any > > experience of how easy, hard or accurate this bit of SA is. I run it at > > work, but the setup precludes training the Bayesian bit. > > I don't really rely on bayesian filters at all, I let the DNS blocklists do > the power filtering and just let the body filters catch the stuff coming from > brand new sources. I don't have particular numbers but on average, I get 800 > spam mails per week through my system and of those, maybe 1 false negative > hits the inbox. > > -- > Bryan Phinney > Software Test Engineer > I took the liberty of adding Bryan's DNS Blacklists to my Postfix server guide if anyone wants to know how to use them. They are really effective at catching a lot of spam. Probably because of the number of times I post on this list, I receive about 400 spams a day, and about 3 a week manage to get through to my Inbox. I have **never** had a false positive! Those messages that do make it to my inbox are generally so obfuscated that they are almost unreadable. derek derek
Want to buy your Pack or Services from MandrakeSoft? Go to http://www.mandrakestore.com
