Am Freitag, 11. Februar 2005 08:14 schrieb Sharma, Ravi: > Its require when you rebuild the database.
That's not true. See the message from Matt. Bayes is a statistical methode so the best way is to train all spam and ham. But you should not train it on messages you whitelist for example with procmail. This would lead to a false database. Thomas > Ravi > > -----Original Message----- > From: Matias Lopez Bergero [mailto:[EMAIL PROTECTED] > Sent: Friday, February 11, 2005 3:36 AM > To: users@spamassassin.apache.org > Subject: bayesian filter training > > Hi > > Just a question, > It is worth to train the bayes filter with messages already detected and > flagged as spam by spamassassin? That would do any good? > > BR, > Matías. > [..] -- icq:133073900 http://www.t-arend.de
pgpA9jFZWP0U9.pgp
Description: PGP signature
