I'm no expert but perhaps you could look at the problem from the other
way around. Maybe you could have 5 or 6 Bayesian databases and feed the
emails as spam or ham to each in turn. Then each filter is saying
yes/no to whether an email is or isn't in one particular category.
Don't forget you have to train each filter about what *isn't* a matching
result as well as what is to get better responses.
But to answer your question I don't think there is anything 'out of the
box' that will do this for you.
- David.
Andoni OConchubhair wrote:
Hi,
I have a need for a categorisation engine for mails coming into my
company to be routed to different queues. I know James has a Baysian
analysis tool for filtering spam but is there any way I could use/train
it to filter into more than just "Yes" and "No". Could I train it to
categorise my emails into 5 or 6 different categories and then tag them.
If I could have the mails arriving in my email application with a tag,
be it a header or a subject line tag of some sort I could then place
them in queues for the relevant person to work on.
Please let me know if this exists for James at the moment.
Regards,
Andoni.
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
