That is what I was beginning to suspect. Is there a way to untrain the emails I ran through it? It was a pretty large selection.. A few thousand of both spam and ham.
I turned on auto-learning so it should start to pick things up on it's own without needing me to train it, no? Training on a per user basis could be difficult as this is a gateway scanning and feeding mail on to around ten or so domains some of which are rather large such as the Caddo Parish School Board that puts through around 10,000 mails per day. Best regards, JD Smith -----Original Message----- From: Sander Holthaus [mailto:[EMAIL PROTECTED] Sent: Monday, April 10, 2006 9:51 AM To: JD Smith Cc: users@spamassassin.apache.org Subject: Re: SpamAssassin Woes -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 JD Smith wrote: > Greetings List: > > My name is JD Smith and I have been put in charge of setting up a spam > solution for my organization. I have chosen to go with MailScanner + > Postfix + SA + MailWatch. > > I have everything pretty much setup and it is working, however my spam > filtering is far from the 90th percentile.. I think I'm actually only > catching around 70% or something which is worse than our old solution. > > I trained the bayes with a corpus of common spam that was recommended to > me by someone somewhere (I forget) when I first got started. Maybe I > need new updated rules? Does anyone have any suggestions on where I > might find a list of good, suggested rules to implement? > > Best regards, > > JD Smith > > Training Bayes with common spam is not the best way. It should be learned with Spam that is specific for your mail-accounts (or better, on a per account basis). SARE has some very good rules. The SpamAssassin Wiki should help you out further. Kind Regards, Sander Holthaus -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.2.2 (MingW32) iD8DBQFEOnDiVf373DysOTURAgIFAJ47fscgLgPAWiCYHmQC3JkEHhMRgQCghOUU ow7Vf7Ho9UZytJt41kJOCRs= =iG6f -----END PGP SIGNATURE----- -- This message has been scanned for viruses and dangerous content by ShooSpam, and is believed to be clean.
