[SAtalk] Site wide Bayes effectiveness with minimal training

2003-08-19 Thread Clayton, Nik [IT]
Hi, Does anyone have any evidence (empircal or anecdotal) as to how well 2.5x fares against 2.4x using Bayes sitewide with minimal training (no more than SA does when autolearning)? It's my understanding (BICBW) that most of the PHRASE rules disappeared in 2.5x in favour of Bayes. Which is all w

Re: [SAtalk] site wide BAYES

2003-06-01 Thread guenther
On Sat, 2003-05-31 at 22:30, William Stearns wrote: > Good evening, Guenther, Good morning, William, > > How can I do that? Did not find any note, how to define a site wide > > BAYES database. TIA > > http://www.stearns.org/doc/spamassassin-setup.current.html#autoreporting Now that page looks

Re: [SAtalk] site wide BAYES

2003-06-01 Thread William Stearns
Good evening, Guenther, On 30 May 2003, guenther wrote: > How can I do that? Did not find any note, how to define a site wide > BAYES database. TIA http://www.stearns.org/doc/spamassassin-setup.current.html#autoreporting Cheers, - Bill --

Re: [SAtalk] site wide BAYES

2003-06-01 Thread guenther
> >Would it be wise (actually, would it do any harm) to have one BAYES > >database for a whole site, instead of per-user basis? (Just a hand full > >of users.) > > There are some advantages and some drawbacks here. On the advantage side as > it learns from the spam of one user it can apply those

Re: [SAtalk] site wide BAYES

2003-05-30 Thread Matt Kettler
At 01:45 AM 5/30/2003 +0200, guenther wrote: hey folks, Would it be wise (actually, would it do any harm) to have one BAYES database for a whole site, instead of per-user basis? (Just a hand full of users.) There are some advantages and some drawbacks here. On the advantage side as it learns from

[SAtalk] site wide BAYES

2003-05-30 Thread guenther
hey folks, Would it be wise (actually, would it do any harm) to have one BAYES database for a whole site, instead of per-user basis? (Just a hand full of users.) How can I do that? Did not find any note, how to define a site wide BAYES database. TIA ...guenther -- char *t="[EMAIL PROTECTED]";