Hello,
[ISO-8859-2] Bolla Peter writes:
Hi,
I've been finding it getting better and better as the training has
started to work. It's taken longer than I expected, but they have had a
lot of variation in their spam, so I guess it's understandable.
What training mode do you use? For me it doesn't show any progress...
After spending some time studying the IP addresses, most (in my case 98%,
YMMV though) originate from dynamic IP addresses. I've had a hard time
getting a good sample, DynaStop has elimited a good majority of them. I
might see 1 or 2 every few days make it through DynaStop. My tests show a
remarkable difference if using the X-DynaStop tag to train on spam.
---
DynaStop: Stopping spam one dynamic IP address at a time.
http://tanaya.net/DynaStop/
