kudret> I'm not keeping my spam emails (I don't see any point doing
kudret> this), permanently delete all of them.
You don't need to keep them. Train as spam, then delete.
kudret> Besides, my 2nd question was "how can get a spammer list?" In
kudret> another words, instead of waiting for weeks to build a spam
kudret> database, there has to be a table somewhere to download and save
kudret> a lot of time.
Sorry, I didn't see that. I generally have to wait about 2 hours for
sufficient spams. Today's not quite over yet and I've received about 250
spams. If you have to wait weeks to receive enough spams to offset the hams
you're training on, then I recommend you simply uninstall SpamBayes. You
don't have a spam problem. ;-)
Seriously, start from scratch. Everything will score as unsure. As new
messages arrive, only train on those messages which are clearly mistakes
(spam scored as ham or ham scored as spam) or those which score as unsure.
Keep your ham and spam thresholds at the defaults, at least to start
(something like 0.2 and 0.7 I think). Don't train on any messages which are
properly classified. Try that for a bit and see if performance improves.
kudret> If you look at my 2 sample reports, both doesn't have any
kudret> problem with original email part. At the bottom, I forgot to
kudret> mention this, you will see AVG virus check report. This part
kudret> somehow shows some spam tokens. First of all, this should not
kudret> be considered as spam and 2nd both reports look the same, should
kudret> not be so different results. And 3rd, like I said at the
kudret> beginning, if my friend sends me 2 emails in same day. When I
kudret> check my emails, they are going to junk folder. This does not
kudret> make any sense.
Sorry, I no longer have your previous message to refer to, so I can't really
recall what your first, second or third messages are. I still think your
imbalanced training is going to create problems.
Skip
_______________________________________________
SpamBayes@python.org
http://mail.python.org/mailman/listinfo/spambayes
Info/Unsubscribe: http://mail.python.org/mailman/listinfo/spambayes
Check the FAQ before asking: http://spambayes.sf.net/faq.html
