From: Jason Gurtz [mailto:[EMAIL PROTECTED]
> > I was curious to know if anyone else has been playing around with
> > subject-line blocking, as discussed in the thread a week or so ago.
> I haven't, but am interested. I'm just curious about what kind of false
> positive rate might be experience
Cormack, Ken wrote:
> My users report any spam that slips through, to an Exchange public folder
> called "spam". I then run a series of VB scripts against any such reported
> spam, to extract relay Ips, subject lines, and other such things that I can
> add to my various rules or databases.
It se
From: David F. Skoll [mailto:[EMAIL PROTECTED]
> It seems to me it would be far more productive and less labour-intensive
> to simply feed those messages into a Bayesian analyzer.
That may be the case. However, the VB scripts have done a pretty good job
so far, are well known and understood by
Cormack, Ken wrote:
> Would a Bayes analyzer deal with a spammer's efforts to "personalize" or
> vary a subject line, with things such as:
> "Hey [EMAIL PROTECTED], claim your dream vacation - 07/15/05"
> "Hey [EMAIL PROTECTED], claim your dream vacation - #GGFS85F"
> ...any more effectively tha
4 matches
Mail list logo