You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@spamassassin.apache.org by Daniel Quinlan <qu...@pathname.com> on 2005/04/27 11:06:12 UTC
Bayes scores for 3.0.3
I propose we change this:
score BAYES_50 0 0 1.567 0.001
score BAYES_60 0 0 3.515 0.372
score BAYES_80 0 0 3.608 2.087
score BAYES_95 0 0 3.514 2.063
score BAYES_99 0 0 4.070 1.886
to
score BAYES_50 0 0 1.567 0.001
score BAYES_60 0 0 3.515 1.0
score BAYES_80 0 0 3.608 2.0
score BAYES_95 0 0 3.514 3.0
score BAYES_99 0 0 4.070 3.5
trivial enough?
--
Daniel Quinlan
http://www.pathname.com/~quinlan/
Re: Bayes scores for 3.0.3
Posted by Loren Wilton <lw...@earthlink.net>.
I'd certainly vote for that, if I had a vote!
Loren
Re: Bayes scores for 3.0.3
Posted by Warren Togami <wt...@redhat.com>.
Daniel Quinlan wrote:
> I propose we change this:
>
> score BAYES_50 0 0 1.567 0.001
> score BAYES_60 0 0 3.515 0.372
> score BAYES_80 0 0 3.608 2.087
> score BAYES_95 0 0 3.514 2.063
> score BAYES_99 0 0 4.070 1.886
>
> to
>
> score BAYES_50 0 0 1.567 0.001
> score BAYES_60 0 0 3.515 1.0
> score BAYES_80 0 0 3.608 2.0
> score BAYES_95 0 0 3.514 3.0
> score BAYES_99 0 0 4.070 3.5
>
> trivial enough?
>
+1, except my gut feeling is that BAYES_60 should be lower to be safe.
Maybe 0.5?
Warren Togami
wtogami@redhat.com
Re: Bayes scores for 3.0.3
Posted by Michael Parker <pa...@pobox.com>.
On Wed, Apr 27, 2005 at 02:06:12AM -0700, Daniel Quinlan wrote:
> I propose we change this:
>
> score BAYES_50 0 0 1.567 0.001
> score BAYES_60 0 0 3.515 0.372
> score BAYES_80 0 0 3.608 2.087
> score BAYES_95 0 0 3.514 2.063
> score BAYES_99 0 0 4.070 1.886
>
> to
>
> score BAYES_50 0 0 1.567 0.001
> score BAYES_60 0 0 3.515 1.0
> score BAYES_80 0 0 3.608 2.0
> score BAYES_95 0 0 3.514 3.0
> score BAYES_99 0 0 4.070 3.5
>
> trivial enough?
+1, I've been running with similar scores for awhile now.
Michael
Re: Bayes scores for 3.0.3
Posted by Duncan Findlay <du...@debian.org>.
On Wed, Apr 27, 2005 at 03:26:14PM -0400, Theo Van Dinter wrote:
> On Wed, Apr 27, 2005 at 12:19:18PM -0700, Dan Quinlan wrote:
> > I think those scores are slightly wrong because we didn't use real-time
> > network results for 3.0.3. The network rule hit rates were too high and
> > the Bayes scores were lowered too far.
>
> Arguably, then, shouldn't we do a new score generation run for 3.0.3?
Please, no!
Anyways, +1 on original proposal to modify Bayes scores.
--
Duncan Findlay
Re: Bayes scores for 3.0.3
Posted by Daniel Quinlan <qu...@pathname.com>.
Theo Van Dinter <fe...@kluge.net> writes:
> Arguably, then, shouldn't we do a new score generation run for 3.0.3?
As long as hell has frozen over, I have no problem with that.
Daniel
--
Daniel Quinlan
http://www.pathname.com/~quinlan/
Re: Bayes scores for 3.0.3
Posted by Theo Van Dinter <fe...@kluge.net>.
On Wed, Apr 27, 2005 at 12:19:18PM -0700, Dan Quinlan wrote:
> I think those scores are slightly wrong because we didn't use real-time
> network results for 3.0.3. The network rule hit rates were too high and
> the Bayes scores were lowered too far.
Arguably, then, shouldn't we do a new score generation run for 3.0.3?
--
Randomly Generated Tagline:
"I'm convinced that the body metal for this car was supplied by Reynold's
Aluminum." - Unknown about the Renault LeCar
Re: Bayes scores for 3.0.3
Posted by Daniel Quinlan <qu...@pathname.com>.
Duncan Findlay <du...@debian.org> writes:
> No, not without a reason. :-)
>
> Are you saying you want to change this since the scores are a bit of
> an anomoly exasperated by the use of old data? The scores were
> probably fine at the time but since the data is now old, it's just not
> right?
>
> +0
I think those scores are slightly wrong because we didn't use real-time
network results for 3.0.3. The network rule hit rates were too high and
the Bayes scores were lowered too far.
Daniel
--
Daniel Quinlan
http://www.pathname.com/~quinlan/
Re: Bayes scores for 3.0.3
Posted by Duncan Findlay <du...@debian.org>.
On Wed, Apr 27, 2005 at 02:06:12AM -0700, Daniel Quinlan wrote:
> I propose we change this:
>
> score BAYES_50 0 0 1.567 0.001
> score BAYES_60 0 0 3.515 0.372
> score BAYES_80 0 0 3.608 2.087
> score BAYES_95 0 0 3.514 2.063
> score BAYES_99 0 0 4.070 1.886
>
> to
>
> score BAYES_50 0 0 1.567 0.001
> score BAYES_60 0 0 3.515 1.0
> score BAYES_80 0 0 3.608 2.0
> score BAYES_95 0 0 3.514 3.0
> score BAYES_99 0 0 4.070 3.5
>
> trivial enough?
No, not without a reason. :-)
Are you saying you want to change this since the scores are a bit of
an anomoly exasperated by the use of old data? The scores were
probably fine at the time but since the data is now old, it's just not
right?
+0
--
Duncan Findlay