|
Sorry...should've mentioned our weighting scale. We
hold at 100 and delete at 300.
Those were just examples, however. The point
is to weight them exactly opposite of your current weight for those
tests.
Darin. ----- Original Message -----
From: Greg
Birdsall
Sent: Thursday, April 14, 2005 5:54 PM
Subject: RE: [Declude.JunkMail] Negative weighting filters to reduce
false positives This is pretty
interesting, but one question – What is your hold weight set to? It seems that you are assigning a huge
negative value for the first test, and much smaller for the other two, nay
insight as to how you came up with these values? We are running into some of the same
problems here, and this is an interesting idea for a way around
it. -
greg From:
[EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]
On Behalf Of Darin Cox We just started something I've been
thinking about for a while: Negative weight tests to offset specific test
failures for well-known domains. For example, a large number of false
positives we see are from Earthlink, Mindspring, Sprint, Verizon,
etc. Now you may be thinking, of course,
these are large providers with dial-up user bases, so you would expect a large
percentage of false positives to be from them...but hold on a
minute. Many of these large domains are being penalized in our system for
routing or not having abuse@ or postmaster@ addresses. Almost all of these
would not have ended up in the hold queue if they had not been so
penalized...thus the idea to figure out a manageable way to NOT penalize them
for these technical RFC violations. So, what we've done is to start
filters to counteract the weights for major tests that a few of these domains
fail. By doing it specifically for a particular domain, we reduce false
positives but avoid losing the effectiveness of the test on other
domains. Anyway, attached zip are the
filter files. As I mentioned, they have just been started, so there are
just a few domains in them at present. At the top of the filter file are
suggested guidelines on how to use them. There are probably better ways to
handle this, so I welcome comments/feedback.
|
- Re: [Declude.JunkMail] Negative weighting filters to reduce fa... Darin Cox
