-----BEGIN PGP SIGNED MESSAGE-----
I have spamd running, and SpamAssassin is catching a good amount of the spam.
I'd like to start training the Bayesian filter and have 2 questions.
(1) spamd is running as root. I know this means that the message filters are
global (no problem, desktop system). I'm not sure what I need to do to use
the Bayesian filters in this case, however. I'd guess that I run sa-learn
against my spam / ham mailboxes, but is there anything else I need to do?
(2) How often should I run sa-learn. I'm saving the spam / ham as it comes in.
I'm assuming that after I run the messages against sa-learn that I should
delete them and collect a new set to expand the training. I'm not sure how
often this should be run, however.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.2-rc1-SuSE (GNU/Linux)
-----END PGP SIGNATURE-----