Craig> The section at the end is simply those rules where there wasn't
Craig> enough data in the corpus to justify allowing the GA to modify
Craig> the scores. Doesn't mean that those rules don't show up at all;
Craig> just that they don't show up more than a small handful of times.
Craig,
Can you give a brief description of the corpus (how big (both spam and
non-spam), how well exercised the rules are, etc)? I would find it useful
to see how many times each rule is triggered. That would allow me to poke
around through my private "corpus" and send you examples that complement
what you already have. For example, I do some ebay stuff from time-to-time.
I could probably save up a few more "question for seller" sorts of mails and
forward them along.
Aside from longer execution time, what's the downside of disabling network
checks in the mass-check script?
(As for the auto-whitelist comment - spamd with the -a flag still isn't
working for me, so that's not really an option. I've been away from mail
for a few days though, so I still have a little sa-talk mail to dig
through. Perhaps the solution lies within...)
--
Skip Montanaro ([EMAIL PROTECTED] - http://www.mojam.com/)
"Excellant Written and Communications Skills required" - seen on chi.jobs
_______________________________________________________________
Don't miss the 2002 Sprint PCS Application Developer's Conference
August 25-28 in Las Vegas -- http://devcon.sprintpcs.com/adp/index.cfm
_______________________________________________
Spamassassin-talk mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/spamassassin-talk
