Spam in images
John G Walker
johngwalker at tiscali.co.uk
Mon Aug 7 22:27:24 CEST 2006
On Mon, 07 Aug 2006 12:31:47 -0700 Bill Wohler <wohler at newt.com> wrote:
>
> Glad to hear it. I'm still getting a dozen unsures every day. I'm
> guessing it takes a few hundred to train properly. Oh wait! Looking at
> my spam folder, it does appear that a lot of these messages are
> getting tagged as spam after all. But there are so many of them!
Can't comment on bogofilter, since I started with a couple of thousand
presorted spam and ham messages to train on, but my first Bayesian
filter - SpamBayes - only needed about thirty or forty messages before
it began to get the hang of my mail.
I suspect any Bayesian filter will learn equally fast. I suspect the
speed is in the mathematics, not the software,
--
All the best,
John
More information about the Bogofilter
mailing list