Hi,
On 03/20/2015 06:50 AM, Reindl Harald wrote:
Am 20.03.2015 um 11:40 schrieb Matus UHLAR - fantomas:
On 20.03.15 09:30, Reindl Harald wrote:
why would you want poems or cooking recipes trained as spam?
and why not?
i think i have explained it often enough now
I've heard arguments in the past for doing it both ways.
Can someone with some authority (ie, those that wrote it or understand
the code) add some input?
Is there no definitive documentation on how the implementation of bayes
in spamassassin works?
Although we may all receive an occasional large message, I don't think
it's the norm for any of us. Bayes poisoning attempts are also much more
frequent - too frequent to have to consider each one before training.
It shouldn't be so controversial.
I now have a spam honeypot that's capturing quite a bit of spam per day
on a completely different network than my other customers, and would
like to know if I can use that data to train bayes?
I'm now trying to figure out how best to use those emails with regards
to bayes. I'd like to be able to share this information with the bayes
database on my production systems.
I'd appreciate any input on what the effects of doing this may be.
Effects of training on the headers? Body? Attachments?
Thanks,
Alex
