SA 2.5 had its first miss for me today. It was a 'you are watching this thread' from The Edge website... OK they deserved it (there was no 'To:', it had faked Outlook when it clearly wasn't, etc.

By the time I caught it, there had been a couple of dozen of these (active thread, presumably). I submitted a couple of the emails to sa-learn-nonspam and it seemed to learn from them, except they only went down to BAYES_44. Since I had specifically said that these emails in particular were *not* spam shouldn't it have gone a lot lower?

I believe SA submits its results back into the bayes system for learning.. does this mean that every time you have a false positive you can only ever cancel it out to zero, never nonspam (Spam report + Nonspam report = 0?).

Tony




-------------------------------------------------------
This SF.net email is sponsored by: SlickEdit Inc. Develop an edge.
The most comprehensive and flexible code editor you can use.
Code faster. C/C++, C#, Java, HTML, XML, many more. FREE 30-Day Trial.
www.slickedit.com/sourceforge
_______________________________________________
Spamassassin-talk mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/spamassassin-talk

Reply via email to