Spammers circumvent MS

[Kevin Spicer]

>Doesn't SpamAssassin (Bayes) create a DB Hash of tokenized results
>(spam and not spam).   I suppose if you could find a way to perform
> matches against that - well, you'd probably re-invent the wheel.

Yes it does, but it only auto-learns very high and very low scoring
mails.  Odds are that the mails your users would want to train it with
would be more moderate scores, as most false positives/ negatives tend
to be borderline scores.  Therefore this wouldn't work unless you
changed the thresholds to tokenise everything, which would probably not
be a good idea (mis-learning borderline cases, server load, database
size etc).






BMRB International
http://www.bmrb.co.uk
+44 (0)20 8566 5000
_________________________________________________________________
This message (and any attachment) is intended only for the
recipient and may contain confidential and/or privileged
material.  If you have received this in error, please contact the
sender and delete this message immediately.  Disclosure, copying
or other action taken in respect of this email or in
reliance on it is prohibited.  BMRB International Limited
accepts no liability in relation to any personal emails, or
content of any email which does not directly relate to our
business.