> I wonder - lately I am getting lots of messages, that only cotain > an image - > which contains the whole message. > Message usually has subject like this: > fw: <nick> > and no text inside (thus poor score possibility) > > Is there any way of teaching SBayes to recognise those as spam ? > They are > unsure usually.
At the moment, the only tokens SpamBayes will generate for such messages are from the message headers. It's possible that this will be enough (with enough training) to learn to identify the messages as spam. I (and no-doubt others outside of the SpamBayes project) am working on improving tokenization of images. Unfortunately, while I have a bit more time to work on this than on getting a 1.1a2 release done, it's still not a lot. Hopefully, I'll be able to get something done that does appear to help, and have it as an experimental option in 1.1a2 (perhaps the end of next month). =Tony.Meyer -- Please always include the list (spambayes at python.org) in your replies (reply-all), and please don't send me personal mail about SpamBayes. http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this. _______________________________________________ [email protected] http://mail.python.org/mailman/listinfo/spambayes Check the FAQ before asking: http://spambayes.sf.net/faq.html
