I was pre-pending a spam text string on the subject line based on the Baysian spam analysis. It was reliably pre-pending my text string. Then about an hour after I first started using this, I started getting an additional "[99.3% SPAM]" or similar text string appended as well. I figure the Baysian mailet is adding it. But I'm really confused why I got a bunch of spams that the mailet caught, but did not add the subject text. I haven't gotten any spams recently that do not have the additional string. But I'd sure like to know what made it start (and what might make it stop again).
I'll remove my subject text if I can figure out when the Baysian mailet does it and when it does not. Where does that "[nn.n% SPAM]" subject text get appended? Thx. Jerry -----Original Message----- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Danny Angus Sent: Tuesday, January 09, 2007 5:48 PM To: James Users List Subject: Re: Best way to feed ham or spam to BayesianAnalysis mailet? On 1/9/07, JWM <[EMAIL PROTECTED]> wrote: > One other question... what happens if I make a mistake and send a ham to the > spam address? How do you undo a mistake in the training? send the same mail to the ham address which will sort of "neutralise" it --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
