[Spambayes] Mass Distribution for Training Set

Tim Peters tim.one at comcast.net
Mon May 24 20:53:32 EDT 2004

[Bahman Lashgari]
> We are considering providing this plug-in to the entire office. 
> However, it is an extra overhead of teaching people how to run training 
> sets

That may be a good reason to consider a different spam filter -- ours is
intended to be personal, and you can't get personal without personal

> and they may not have enough emails for the spam category to build 
> a good and updated set.

It can do surprisingly well with just a few dozen of each.  If your folks
don't have a few dozen spam, save everyone a lot of bother and don't install
a spam filter at all <0.9 wink>.

> Our question is this: can we configure one  training file and load
> the same training file on all machines as default set?  In this case,
> for example, the training file would be training.file and we could
> copy and paste to all workstations.

There's no technical problem in doing that.

> How would this work?

I'm not aware of any research on this, so can't say.  If you try it, let us
know how it works!  The commercial SpamBayes derivative here:


apparently comes with some pre-training, but we're not privy to the details
of how they did that, or of how well it works for their user.

More information about the Spambayes mailing list