reverse MCP

[Eric Dantan Rzewnicki]

On Wed, Sep 08, 2004 at 10:55:11PM +0200, Remco Barendse wrote:
> On Wed, 8 Sep 2004, Julian Field wrote:
> >At 21:24 08/09/2004, you wrote:
> >>On Wed, Sep 08, 2004 at 09:38:55PM +0200, Remco Barendse wrote:
> >>>Why would you want to do that?
>
> Would bayes be the suitable tool for that? If you are interested in
> kernel+scsi stuff this month but next month you decide that sata stuff is
> more interesting you end up with mails about scsi stuff stil scoring high.

That's true, I guess. Hadn't considered that. Good point.

> I very much like the idea of this but guess that statically assigned rules
> will be better? Unless your favourite topics never change ofcourse. :)

Well, I guess over time with training, you would sort of be punished for
your changing interests by a lag in bayes response. :) probably static
rules would be best, as you suggest.

-Eric Rz.

------------------------ MailScanner list ------------------------
To unsubscribe, email jiscmail at jiscmail.ac.uk with the words:
'leave mailscanner' in the body of the email.
Before posting, read the MAQ (http://www.mailscanner.biz/maq/) and
the archives (http://www.jiscmail.ac.uk/lists/mailscanner.html).