I recently got a false positive largely due to this rule:
* 3.0 SUBJ_YOUR_DEBT Subject contains "Your Bills" or similar
I think it's interesting that a rule can get scored so high when it hits
more ham than spam:
MSECS SPAM% HAM% S/O RANK SCORE NAME WHO/AGE
0 0.0115 0.0306 0.273 0.43 3.30 SUBJ_YOUR_DEBT
header SUBJ_YOUR_DEBT Subject =~ /Your (?:Bills|Debt|Credit)/i
score SUBJ_YOUR_DEBT 3.299 3.045 1.199 0.987
I can certainly imagine how it's possible, just wondering if it should be
improved. Maybe the rescorer should limit scores... without more evidence
against them, or something?
My credit card company provides me with a service that emails me once a
month to let me know I have a new credit report. It's really ham.
--
"Just because you're offended, doesn't mean you're right." - Ricky Gervais
http://www.ChaosReigns.com
