Re: Learning/relearning twice; effect on database

[email protected] said:
> Ok, that makes sense.  How large is "large" - 100, 1000, or 10000? 

About 200, I'd say.

[email protected] said:
> I move messages manually when they have been miscategorised.  Then
> nightly I scan folders for messages with ifile headers which don't
> match the folder they're in - these are messages that have been
> miscategorised.  I --delete these messages from ifile-categorised
> folder and --insert into folder they're sitting in.  What you're
> saying sounds like I should keep track of which messages I have
> recategorised in this way, as over time misclassified ones will be
> overweighted if I process them repeatedly..   

I see what you're doing here; sounds like you're doing The Right Thing.  
When else do you call ifile?  Do you use --query-insert when 
incorporating new e-mail?

Ah, yes.  I understand now.  Your intuition is right.  You should keep 
track of those messages that you've recategorized on previous runs.  In 
your case, it might make the sense to modify the X-filter: header to 
indicate the new folder, though that header is useful for keeping track 
of how well ifile does it's job.

Jason D. M. Rennie
MIT AI Lab
[email protected]
(617) 253-5339
http://www.ai.mit.edu/~jrennie/



------------------------ Yahoo! Groups Sponsor ---------------------~-->
Get 128 Bit SSL Encryption!
http://us.click.yahoo.com/JjlUgA/vN2EAA/kG8FAA/CefplB/TM
---------------------------------------------------------------------~->

To UNSUBSCRIBE, send an empty message to
[email protected]

ifile web site: http://www.ai.mit.edu/~jrennie/ifile/ 

Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/