[Spambayes] Basic Stuff

[Katz, Amir]

AFAIK, if you don't make any changes to the results (that is, SB was
correct), training is pretty fast. You can probably set the default radio
button to be 'discard' on the 'ham' and 'spam' and just worry about the
messages that were categorized as 'unsure'. However, maybe (and I'm sure the
developers can answer that) when a message that was categorized as spam has
the radio button set to 'spam', it has the same effect as 'discard'.
 
IMO, the only activity that takes noticeable time is when you click on
'review messages' and then the web server collects the untrained messages
and displays them, one screen per day (or maybe more). However, I get my
kicks from watching all the spams intercepted, so the wait time is well
spent :-)
 
Amir 
 

  _____  

From: Philip Delancy [mailto:philip.delancy at tesco.net] 
Sent: Monday, September 27, 2004 14:21
To: Spambayes Mailing List
Subject: [Spambayes] Basic Stuff


I run Outlook Express 6 and my Spambayes installation seems to work fine.
It is particularly good at not classifying "ham" as "spam".  Basic
questions:
Do I have to keep training all the messages that keep coming in?  If I'm
away and come back to 100 or so, this can take time.  What happens to the
messages if I don't train them?
 
The way I operate Spambayes is to review all messages, reclassify a few and
delete from the OE6 Inbox any that have slipped through.
 
If that is normal, then that probably saves me a bit of time.  Is there
anything else I should be doing, with the Clues/Tokens thing for example?
 
Otherwise, keep up the good work!
 
Phil Appleton
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/spambayes/attachments/20040927/1b8b3a48/attachment.html