I upgraded from 1.3.5(final) to 1.3.5(18.2), now (18.9). since the move
from (final) I noticed a more than tripled increase in reported missed
spam and whitelist submissions. I took a look today at the
rebuildrun.txt to see if the issue might be Bayesian and the norm is now
at 1.0488. Prior to the move to 18.x, it was consistently at .59x or
.60x. My understanding is the closer to .6 it is, the more accurate the
bayesian filter will be. Any suggestions on how to 'fix' it? I am
using the latest rebuild script from fritz's dev site.
Found 12836215 spam words, 12238641 non-spam words.
Generating weighted keys...
norm=1.0488
Thu Mar 27 07:54:39 2008: Saving rebuilt SPAM database
Thu Mar 27 07:54:43 2008: Analyzing ended
spam.db: 8708096 bytes
Paul K. Dickson
Systems Administrator III
Frederick County Government, I.I.T.
301-600-2399
-------------------------------------------------------------------------
Check out the new SourceForge.net Marketplace.
It's the best place to buy or sell services for
just about anything Open Source.
http://ad.doubleclick.net/clk;164216239;13503038;w?http://sf.net/marketplace
_______________________________________________
Assp-test mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/assp-test
