well ham is very site dependant (point taken especially with security reasons), so i dont think sending ham will really be something that can be accomplished very easily so i think sticking to spam only would be best. By only sending spam, i would hope to achieve a large spam corpus that someone can incorporate into their own already collected and trained ham corpus. Some people just dont get enough spam, and need a good amount of spam to get a good trained bayes. That is where i would like to help.
I do see the point of the special X-headers that get inserted into mail and can be wrongly learned as spam. I also agree with the fact that spam is very site dependant, and spam to one may be ham to another, but maybe if we established some guidelines as to what kind of spam could be entered, as well as some header rules , we can achieve something. Razor does it with very good success so i think we too can conjure something together. I am just trying to give back to the SA community for all the hard work they have put in. Everyone here at this point has a really good experience with SA so i figure with all of our heads put together we can possibly come up with a good supplement to help SA be even better at what it already does so well. adam On Thu, 2003-12-11 at 09:46, Fred wrote: > Adam Denenberg wrote: > > SA List, > > > > What i want to start is a Bayes Corpus Project. I would like to be > > able to allow people to submit confirmed ham and/or spam to a large > > bayes corpus repository (or maybe just spam) where people could then > > download (or somehow do an sa-learn remotely) to an ongoing updated > > bayes corpus. > > > > > > Feedback and ideas welcome and appreciated. > > > > thanks > > adam > > Only sending spam could be a bad thing. > Example, say my SMTP daemon inserts a special header (X-Foo) into all my > mail (spam & ham). If I submit only my spam to your corpus, every's bayes > system will think that mail with (X-Foo) header is spam. > > The remote sa-learn part could be automated using wget and a cron job. > > ------------------------------------------------------- This SF.net email is sponsored by: IBM Linux Tutorials. Become an expert in LINUX or just sharpen your skills. Sign up for IBM's Free Linux Tutorials. Learn everything from the bash shell to sys admin. Click now! http://ads.osdn.com/?ad_id=1278&alloc_id=3371&op=click _______________________________________________ Spamassassin-talk mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/spamassassin-talk
