You are viewing a plain text version of this content. The canonical link for it is here.
Posted to users@spamassassin.apache.org by Miguel <mm...@123.com.sv> on 2007/09/07 20:29:15 UTC
newbie question about bayesian filter
Dear All, im using SA in a central system wide dedicated filter, so i
dont have any account in it, the "clean" emails are forwarded to the
final destination servers.
In this scenario, nobody will be training SA, so, does it make sense to
use the bayesian filter at all? ,or SA will still use the bayesian
filter from the day to day spam/ham hit ratios?
Thanks
---
Miguel
Re: newbie question about bayesian filter
Posted by maillist <ma...@emailacs.com>.
Miguel wrote:
> Dear All, im using SA in a central system wide dedicated filter, so i
> dont have any account in it, the "clean" emails are forwarded to the
> final destination servers.
> In this scenario, nobody will be training SA, so, does it make sense
> to use the bayesian filter at all? ,or SA will still use the bayesian
> filter from the day to day spam/ham hit ratios?
>
> Thanks
> ---
> Miguel
>
I don't know about anyone else, but the bayes system *to me* is the
bread & butter of spamassassin, and *for me* makes the whole process
painless. If I get spam that doesn't score high enough, I run "sa-learn
--spam /path/to/file" and I never see that spam again.
But then again, I use one server, and have less than 100 users, so doing
this manually is not a bad idea for me.