Ed
At 07:33 PM Thursday, 7/31/2003, Sanjay K. Patel wrote -=>
How can someone generate this report on the their local copy of spamassassin?
-SKP
-----Original Message----- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Matt Kettler Sent: Thursday, July 31, 2003 3:42 PM To: Clay Atkins; [EMAIL PROTECTED] Subject: Re: [SAtalk] DCC vs Razor2 vs Pyzor
At 01:45 PM 7/31/2003 -0500, Clay Atkins wrote: >Any thoughts on which of these is better, none, or a combination of them?
The STATISTICS*.txt files that come with spamassassin will speak much more definitively about the accuracy of these tools than any human on this list can.
from SA 2.55's rules/STATISTICS-set1.txt
OVERALL% SPAM% HAM% S/O RANK SCORE NAME 27.427 51.2148 0.1658 0.997 1.00 2.06 RAZOR2_CHECK 13.201 24.7144 0.0074 1.000 0.97 4.40 PYZOR_CHECK 6.254 11.6769 0.0384 0.997 0.94 3.02 DCC_CHECK
Razor seems to have the greatest number of hits, pyzor seems to have the highest accuracy, although all three have good accuracy. DCC seems to be a tie with razor for accuracy, but has a much lower hit rate than either razor or pyzor.
(Note: I'm deriving "accuracy" from the S/O of the rule. The S/O figure is a "what percentage of the hits were spam hits". So the closer S/O is to 1.0, the more accurate the rule is at correctly picking spam without picking up nonspam.)
Ed Kasky Los Angeles, CA . . . . . . . . I would have made this shorter, but I ran out of time. . .
------------------------------------------------------- This SF.Net email sponsored by: Free pre-built ASP.NET sites including Data Reports, E-commerce, Portals, and Forums are available now. Download today and enter to win an XBOX or Visual Studio .NET. http://aspnet.click-url.com/go/psa00100003ave/direct;at.aspnet_072303_01/01 _______________________________________________ Spamassassin-talk mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/spamassassin-talk
