Alexander K. Seewald wrote:
> Best way would be a simple way to get the probability
> of the bayes classifier within SA for each mail - is that possible?

You can use the BAYES_* rule in the X-Spam-Status header. If that is not
fine-grained enough, you can add a header containing the Bayes score
using a configuration option such as

add-header all Bayes-Score _BAYES_

which would add to all mail a header that looks like

X-Spam-Bayes-Score: 0.9633

The string "Bayes-Score" above is an arbitrary one that you provide to
name the header what you want. man Mail::SpamAssassin::Conf has the
details on the add-header directive.

> An incremental feedback mechanism that allows the score set to
> evolve could help

The recent work on getting a more agile mechanism for updating rules in
between releases of SpamAssassin is a big step to addressing the
problems you brought up, but it leaves open the problem of adjusting the
scoring for the more rapidly changing rule set. I think an incremental
feedback mechanism will become even more necessary when we are able to
update rules more frequently.

 -- sidney

Reply via email to