Re: Troubleshooting Bayes

On Thu, 2008-03-13 at 23:28 -0400, Tim Palmer wrote:
> 0.000          0     244301          0  non-token data:
nspam
> 0.000          0      51666          0  non-token data:
nham

Are you by any chance experiencing false positives due to
bayes scoring?
I had that problem when my counts were so lopsided.

-- 
Robert

_______________________________________________
Maia-users mailing list
Maia-usersrenaissoft.com
http://www.renaissoft.com/mailman/listinfo/maia-users

Robert Fitzpatrick wrote:
> On Thu, 2008-03-13 at 23:28 -0400, Tim Palmer wrote:
>   
>> 0.000          0     244301          0  non-token
data: nspam
>> 0.000          0      51666          0  non-token
data: nham
>>     
>
> Are you by any chance experiencing false positives due
to bayes scoring?
> I had that problem when my counts were so lopsided
Hard to say. I'm one step removed from the users on this
server. For my 
account on a domain on this, mailing lists do get FP fairly
often, but I 
don't mind (<shameless plug for Maia> thanks to the
ease of recovery!). 
It is, of course, standard user behavior to not confirm
much, and to 
mostly confirm FPs, without really clearing out ham/spam
caches. I fully 
expect low confirmation rates, and a basis in bayes towards
nspam
_______________________________________________
Maia-users mailing list
Maia-usersrenaissoft.com
http://www.renaissoft.com/mailman/listinfo/maia-users