Is there anything I can do to help improve the detection rate? I'm willingDefinately a good point on the poisoned Bayes. You want to be obsessive about keeping spam, and ham separate - in doubt, leave it out. And obviously the best way to be sure, is train manually.
Installing DCC - http://www.rhyolite.com/anti-spam/dcc/ - will also give great improvement - hits always scores 2.9 in my installation.
Install Razor too - http://razor.sourceforge.net/
Bryan
A little explanation might be in order - I'm measuring against mails that--
I'm getting that are addressed to abot 40 former employees. All emails get
forwarded to our Exchange server, and since I can consider all of these
mails to the former employees to be spam, all I have to do is save all of
those to a separate directory, and analyze them, instead of all of the mails
on the Exchange server.This helps a great deal, since it isn't a measurement against all of the
current employees, which I think would be far too time-consuming to figure
out, given that I'd have to separate out the non-internet mail from the
internet mail, then classify what what caught by spamassassin vs. not, etc.
I'm making the (hopefully warranted) assumption that the spam that those who
are departed is pretty much the same as the spam that those who are
currently employed are receiving.Thanks all,
Kurt Buff
Sr. Network Administrator
Zetron, Inc.
425.820.6363 x463
[EMAIL PROTECTED]
PO Box 97004
Redmond, WA 98073
-------------------------------------------------------
This SF.net email is sponsored by: IBM Linux Tutorials.
Become an expert in LINUX or just sharpen your skills. Sign up for IBM's
Free Linux Tutorials. Learn everything from the bash shell to sys admin.
Click now! http://ads.osdn.com/?ad_id=1278&alloc_id=3371&op=click
Nothing in the world has more potential for beauty than woman. Nothing has more potential to destroy it, than the world. - (Anonymous)
http://www.wecs.com/content.htm
This signature file is generated by Pick-a-Tag !
Written by Jeroen van Vaarsel
http://www.google.com/search?hl=en&ie=ISO-8859-1&q=pick-a-tag
