Anyone?  *taps on microphone* is this thing on?

On Sunday, Feb 16, 2003, at 20:31 US/Pacific, John Rudd wrote:


I have an automated spam handling set up right now which takes messages I put into a certain mbox-folder and uses them to add to my existing blacklist file (and then a cron job creates a prefs file out of that blacklist file and a few other files). That folder is currently named "Blacklist". When my program is done, it deletes the message.

With the coming of the next SA version, with bayesian learning, I want to incorporate that into the mix. I'm thinking I'll have two mbox-folders, one named "False-Positives" and one named "False-Negatives". The "False-Negatives" file will essentially be the same as the existing "Blacklist" file, except that I want it to also put the message through to the bayesian engine for saying "this was supposed to be marked as spam". Similarly, the "False-Positive" folder would send the messages through as "should not have been marked as spam".


The main question is ... do I need to save the messages in any particular place? Do I _need_ to build them up as a corpus for later review? Or can I just get rid of the ones that were spam and deliver back into my other mail folders the ones that weren't?



-------------------------------------------------------
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf
_______________________________________________
Spamassassin-talk mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/spamassassin-talk


-------------------------------------------------------
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