Bayes still effective?

[Raymond Dijkxhoorn]

Hi!

> Thanks for the stats, Raymond.  They mirror mine in terms of RCVD_IN_SBL+XBL
> and RCVD_IN_BL_SPAMCOP_NET being the most effective rbl providers.
>
> Of course, the stats don't show how many spam messages BAYES_99 caught that
> the other methods wouldn't have done.  They don't show either whether
> BAYES_99 was correct.
>
> Tell me, do you teach the Bayes engine spam with the lists of words within
> them?

It is auto learning, and doing a nice job. But then again with 2.5M
messages a day its pretty hard to really poison it. For us currently its
costing a lot of CPU, but still gives the little extra.

Bye,
Raymond.

-------------------------- MailScanner list ----------------------
To leave, send    leave mailscanner    to jiscmail at jiscmail.ac.uk
Before posting, please see the Most Asked Questions at
http://www.mailscanner.biz/maq/     and the archives at
http://www.jiscmail.ac.uk/lists/mailscanner.html