Re: classifier questions

Jason White <[email protected]> Tue, 12 May 2009 13:14:13 +1000
Newsgroups gmane.mail.spam.crm114
Message-ID <[email protected]>
Eric S. Johansson <[email protected]> wrote:
 
> As they say in the classics, "your mileage may vary.".  I've had horrible luck
> and my OSBF score distributions are what a very kind person would call
> "bimodal".  I get a huge blob of scores below zero and a smaller, rounder blob
> of scores above zero.

This is interesting. Even after fixing my MUA configuration to pipe messages
to mailreaver with CRM114 headers intact, and starting with newly initialized
CSS files, after some training, OSBF starts classifying all messages as
"good", with quite high scores.

HyperSpace (which I'm now trying) seems to be better, but after several days
of training, most messages are still marked as "unsure", although the actual
scores appear to be tending in the right direction.

Any suggestions as to how I can get substantially higher accuracy would be
welcome. I am training all errors and unsures as I encounter them.


------------------------------------------------------------------------------
The NEW KODAK i700 Series Scanners deliver under ANY circumstances! Your
production scanning environment may not be a perfect world - but thanks to
Kodak, there's a perfect scanner to get the job done! With the NEW KODAK i700
Series Scanner you'll get full speed at 300 dpi even with all image 
processing features enabled. http://p.sf.net/sfu/kodak-com