SUL19,
hello guys, are my questions understandable or not?
I think I understand your question. This is a low bandwidth list so one reason for a lack of response is there aren't many of us here ;-)
I also think people are reluctant to respond to people who can't be bothered to give their name. Are you (SUL19) related to LUQ for example who also wrote messages on the same subject at the same time?
To answer your question. The normal process of training the spam or ham database is to gradually and manually send emails to the ham or spam address as attachments. You can send many attachments at the same time if you like to speed things up. This is the best way to do it as you allow the bayesian process to learn the 'shape' of your particular spam and more importantly your particular ham.
One short cut is to get someone else's database that they have built up over time and install it on your machine... but that is never very satisfactory because it isn't tailored to the shape of your emails.
For more information about the Bayesian analysis feeder and how to set it up you may like to follow this email thread... I can't be bothered to repeat it here again! http://www.nabble.com/How-do-I-reduce-SPAM-to21763054.html#a21763054
Regards, David Legg --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
