Re: Re: spam / train-spam

Ray wrote:
> On Wed, Jun 22, 2005 at 10:38:12PM -0500, John Chandler wrote:
>>Should the "good" and "spam" commands _ever_ be used?
>>If so, when?
> 
> If you don't mind some disk activity, spam would probably get you
> higher accuracy.  Of course, the difference should be marginal to
> non-existent and more theoretical than practical.

I agree with that analysis.  Use good, spam, and receive if you don't 
care about disk space or I/O and want the highest possible accuracy. 
Use train-good, train-spam, and train if you want to reduce disk space 
consumption and I/O at the loss of only a tiny amount of accuracy.

Personally I haven't used good, spam, and receive for my own email since 
I added the train-good, train-spam, and train commands.

If you use the train commands then you can use auto-train when building 
your initial database since it will attempt to minimize your database size.

It all comes down to giving people options and gentle hints.  I try not 
to tell people what to do but rather to describe the options available 
and gently nudge them towards the ones I prefer.

All the best,
++Brian


-------------------------------------------------------
SF.Net email is sponsored by: Discover Easy Linux Migration Strategies
from IBM. Find simple to follow Roadmaps, straightforward articles,
informative Webcasts and more! Get everything you need to get up to
speed, fast. http://ads.osdn.com/?ad_id=7477&alloc_id=16492&op=click