I just noticed the default collecting option for spam bombs (spamBombLog) is "spam folder & sendallspam". I thought the idea behind catching spam bombs was to prevent them from disrupting the corpus and disrupting the bayesian test. So would it not be safer to discard them? And related to this, tests that fail because of the message score and have a bomb test that was scored are going to the spam corpus. They used to be discard. I'm assuming there is a reason for these to changes.
Also, on the subject of bomb test. The default penalty value for a bombraw (bombValencePB) is 20, default for bayseian (baysValencePB) is 39. So a test that ONLY fails with these two tests is going to be rejected (PenaltyMessageBlock default at 50). IMO it would be *safer* if one more test is required to cause this email to be marked as spam. Both bayesian and bomb test are, in my experience, unreliable. So for me, I've gone ahead and lowered bombraw to 10. Obviously I can change my settings to my needs, but I'm thinking of safe settings (default) for new installations. ------------------------------------------------------------------------------ Open Source Business Conference (OSBC), March 24-25, 2009, San Francisco, CA -OSBC tackles the biggest issue in open source: Open Sourcing the Enterprise -Strategies to boost innovation and cut costs with open source participation -Receive a $600 discount off the registration fee with the source code: SFAD http://p.sf.net/sfu/XcvMzF8H _______________________________________________ Assp-test mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/assp-test
