[Spambayes] SpamBayes v1.04 Windows
bishop at aeroprise.com
Fri Dec 15 19:11:58 CET 2006
I would try looking at the spam clues for a good email.
It almost sounds as if you have been training SpamBayes the wrong way
around. If you see a keyword like: 'from:addr:personwhosenttheemail', and
if the #ham for this is zero and #spam is more than one, then you have been
training it the wrong way.
Other possibilities are:
a) you have not trained SpamBayes on enough emails.
I suggest going to SpamBayes Manager->Training, and re-doing the training,
selecting some substantial folders with good save email, and, if you have
any folders containing only spam, use them in the training. Rebuild the
training database. Then you should be able to immediately go to a good
email and get the spam clues for the email. Also, you can put a column in
your outlook that gives the spam score for each email. If it doesn't show
up for certain emails, this is because somehow SpamBayes is not filtering
I suspect you have set your filter parameters improperly. I find that
concluding things are junk suspects at 15%, and junk certain at 75% is
better. Remember, it is not the goal to minimize the number of emails in
the junk suspects folder, it is simply to limit the junk suspects to those
emails that SpamBayes really needs to be trained on. Then you train it, and
then the volume in the junk suspects folder drops. Setting junk suspects to
45% will put a lot of spam into your inbox. Also, a narrow difference of
45-68 for the junk suspects folder will not properly identify the emails
that SpamBayes needs to be trained on. There will be very few emails in
junk suspects, but you will be getting good emails in spam certain (68 is a
little low for the spam certain folder), and you will be getting lots of
spam in your inbox.
From: spambayes-bounces at python.org [mailto:spambayes-bounces at python.org] On
Behalf Of tedkilroy
Sent: Wednesday, December 13, 2006 12:46 PM
To: spambayes at python.org
Subject: [Spambayes] SpamBayes v1.04 Windows
Am I missing something, all my ham is ending up in junk email folder not in
the junk suspects folder. I set a score of 68 to be known spam and 45 to be
placed in the junk suspects folder. I also used the recover from Spam so
that SpamBAyes learns but all POP3 internal mail is ending up in the junk
email folder instead.
Any suggestions wouls be appreciated.
View this message in context:
Sent from the Python - spambayes mailing list archive at Nabble.com.
SpamBayes at python.org
Check the FAQ before asking: http://spambayes.sf.net/faq.html
More information about the SpamBayes