[Spambayes] Problem with POP3 Proxy: Sorting Accuracy?

Tony Meyer tameyer at ihug.co.nz
Wed May 5 17:40:41 EDT 2004


> After a week, I feel that Spambayes is not sorting 
> accurately. I have trained accurately, but this morning, all 
> of the deferred messages - 30 or 40 - were spam. About 20 of 
> 60 messages sorted as ham were, in fact, spam. The messages 
> sorted as spam are indeed spam, but Spambayes is missing too many.
>  
> Am I doing something wrong? I just saw a post that someone 
> was getting 99% accuracy after only 2 days! Any suggestions 
> you can provide will be greatly appreciated.

If you could provide an example message that was incorrectly classified,
along with the "clues" for the message (on the review page there is a
"Clues" link that will get you this), that would help us a lot.  You might
even be able to see why the message scored what it did yourself, but if you
can't, we should be able to explain.

Note that it's fairly common for the majority of unsure messages to be spam.
Sometimes altering the threshold can fix this problem (if they're all
scoring > 75%, for example).  You should only get ~2-5% of your messages
classified as unsure, though.

=Tony Meyer

---
Please always include the list (spambayes at python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes. This
way, you get everyone's help, and avoid a lack of replies when I'm busy.




More information about the Spambayes mailing list