[Spambayes] Is Equal Ham & Spam really the best?
David Abrahams
dave at boost-consulting.com
Mon Jul 30 02:59:12 CEST 2007
on Sun Jul 29 2007, skip-AT-pobox.com wrote:
> Dave> Even training only on mistakes and unsures, I have had a steadily
> Dave> increasing ratio for months. I almost never see a misclassified
> Dave> ham and only very rarely a ham about which the system is unsure.
> Dave> It's unsure about spam every day.
>
> What are your ham and spam thresholds set to?
The default, whatever that is, for classification. And I've been
using
-o Categorization:ham_cutoff:0.02 \
-o Categorization:spam_cutoff:0.98 \
when I invoke tte.py
> Have you tried lowering your spam threshold a bit?
No, I haven't tried that. I suppose I run the risk of a few false
positives if I go too far with that?
--
Dave Abrahams
Boost Consulting
http://www.boost-consulting.com
The Astoria Seminar ==> http://www.astoriaseminar.com
More information about the SpamBayes
mailing list