Don't worry about the fact that the spam messages are attachements of the report (and that the attachment has no headers). Sa-learn will realise this and strip it out. - at least that's what I've been told, and bayes is working like a treat for me- with a spam corpus of 3300 messages using libpst and PST Export in Outlook together with mostly auto-learned ham.
-----Original Message----- From: Colin A. Bartlett [mailto:[EMAIL PROTECTED] Sent: 30 October 2003 12:17 To: [EMAIL PROTECTED] Subject: [SAtalk] outlook corpus? All, I have amassed a pretty large corpus of spam (over 16,000 messages). Unfortunately they are just in a folder in my Outlook. (Sorry. I like it.) And, they are all attached to the SA-generated report message. I've mined Google and the archives. I've found some references to converting Outlook files to mbox format. That would work I believe except that the messages I want are really attached to those in my Outlook. Anyone know of a tool or method of generating an mbox or other file from my Outlook messages for use in checking rules and such? I'd hate to have to throw out my 16,000 spams as they are pretty varied, hand verified, and would be great for testing. cheers, Colin Colin A. Bartlett Kinetic Web Solutions www.kineticweb.biz ------------------------------------------------------- This SF.net email is sponsored by: SF.net Giveback Program. Does SourceForge.net help you be more productive? Does it help you create better code? SHARE THE LOVE, and help us help YOU! Click Here: http://sourceforge.net/donate/ _______________________________________________ Spamassassin-talk mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/spamassassin-talk ------------------------------------------------------- This SF.net email is sponsored by: SF.net Giveback Program. Does SourceForge.net help you be more productive? Does it help you create better code? SHARE THE LOVE, and help us help YOU! Click Here: http://sourceforge.net/donate/ _______________________________________________ Spamassassin-talk mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/spamassassin-talk
