On Mittwoch, 9. November 2005 08:04 Gary W. Smith wrote:
> My users are quite happy
> with overall markup of the spam. We occasionally get a HAM marked as
> SPAM. We have an odd client base though.
The question is: when to use global and when per-user bayes?
On our server, we have people of different languages, communicating with
different countries all over the world, in different areas
(advertising, production, IT, etc.). I thought in that case a per-user
bayes would be much better, as viagra is something good for the one,
but bad for the other.
What's the general recommendation for bayes?
mfg zmi
--
// Michael Monnerie, Ing.BSc --- it-management Michael Monnerie
// http://zmi.at Tel: 0660/4156531 Linux 2.6.11
// PGP Key: "lynx -source http://zmi.at/zmi2.asc | gpg --import"
// Fingerprint: EB93 ED8A 1DCD BB6C F952 F7F4 3911 B933 7054 5879
// Keyserver: www.keyserver.net Key-ID: 0x70545879
pgpCN6ryXTaZ2.pgp
Description: PGP signature