This message from the Razor list makes me think maybe there'd be some use in
having mass-check record the dates of the emails processed. Then we could turn
scores per weekday (mon,tue,wed,etc), and it'd also be a way of making newer
occurrences be more important than older occurrences, so we'd track trends in
spam, and not worry so much about older patterns which don't occur as much any
more. Any thoughts?
C
---------- Forwarded message ----------
Date: Fri, 12 Apr 2002 09:14:11 -0400
From: Fox <[EMAIL PROTECTED]>
To: [EMAIL PROTECTED]
Subject: [Razor-users] This week in Razor stats - Apr 3-11
Notice Razor does about 35% on the weekends compared to around 20% during the week.
This is because Razor is probably not getting many submissions from business to
business opt-in spammers like Cahnersdirect. SpamJammer does the worst on the
weekends, being better at stopping the b2b spam than the "pure" spam we get on
weekends. They compliment each other nicely.
Each spam email is tested for all four conditions, SpamJammer, Razor, sender
blacklists, and being a trollbox. If you have any questions about the stats, contact
me at [EMAIL PROTECTED]
Spam Statistics
Date Spam Count Jammed Count Razored Count Blacklisted Count Trollbox Count
2002-04-03 194 123 (63.40%) 37 (19.07%) 21 (10.82%) 76 (39.18%)
2002-04-04 149 105 (70.47%) 35 (23.49%) 12 ( 8.05%) 53 (35.57%)
2002-04-05 129 97 (75.19%) 42 (32.56%) 14 (10.85%) 38 (29.46%)
2002-04-06 69 47 (68.12%) 25 (36.23%) 8 (11.59%) 33 (47.83%)
2002-04-07 74 43 (58.11%) 37 (50.00%) 5 ( 6.76%) 35 (47.30%)
2002-04-08 109 82 (75.23%) 29 (26.61%) 10 ( 9.17%) 41 (37.61%)
2002-04-09 194 157 (80.93%) 34 (17.53%) 16 ( 8.25%) 63 (32.47%)
2002-04-10 140 104 (74.29%) 32 (22.86%) 6 ( 4.29%) 64 (45.71%)
2002-04-11 165 143 (86.67%) 26 (15.76%) 36 (21.82%) 48 (29.09%)
Fox
_______________________________________________
Spamassassin-talk mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/spamassassin-talk
