On Tue, 24 Jul 2018 10:34:42 -0400
Kris Deugau wrote:

> Kris Deugau wrote:
> > Nick Bright wrote:
> >   
> >> The problem I'm trying to solve is "how to implement a training
> >> system on my server".  
> > 
> > I'd suggest a manual review step before feeding the messages to
> > Bayes.
> > 
> > You **WILL** get users reporting all kinds of "unwanted today
> > because Reasons" but otherwise legitimate email as spam.  
> 
> I'd also like to note that most of the advice from others who've
> replied looks to me like advice for a corporate mail system or small
> private domain.
> 
> Much of that advice doesn't scale or behave well when applied to ISP 
> mail services of more than a couple hundred users, IME.

IMO service providers should use per-user Bayes. 

Reply via email to