-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
In <mid:[EMAIL PROTECTED]>,
Peter Fjelsten [PF] wrote:'
PF> Actually, I only get false negatives with Bayesian and no false
PF> positives with blacklists.
No false positives with BlackLists? Wow! :)
PF> What is the basis of your statement?
When I first got SpamPal, I chose the intermediate level for spam
detection. I then used a filter to create a copy of all messages
designated as Spam and place the copies in a special folder.
Over 90% of the catches were legitimate mail. A lot were from list
members who were using black-listed SMTP servers. I'm sure that with
Whitelist building etc. the false positive rate would diminish.
I don't use SpamPal in this way anymore. I now filter all my known
mail and then have SpamPal work with what's left. It has been doing
fairly well there.
PF> I have been using SpamPal (with Bayesian) for more than a month now
PF> with only Bayesian giving me problems.
The Bayesian filter plug-in is yet to detect one of my spam
messages. :/
PF> But I unchecked some of the Blacklists (e.g SpamCop's).
I have SpamCop's enabled. With it disabled, SpamPal was missing
Spam.
- --
-=] allie_M [=- {List Moderator}
-
-----BEGIN PGP SIGNATURE-----
Comment: My Public Keys - http://www.ac-martin.com/pgpkeys.html
iD8DBQE+Vr78V8nrYCsHF+IRAm89AKC66AFZtZC0eZ3Zck/7fJQd6noUDACeIrPm
Gr+yGBenIVCznwVJJtSxLRg=
=bpAh
-----END PGP SIGNATURE-----
________________________________________________
Current version is 1.62 | "Using TBUDL" information:
http://www.silverstones.com/thebat/TBUDLInfo.html
