Re: Bayes scores in SA 3.0

[Chip Paswater]

> >  Hey guys,
> >  
> >  I was looking at the Bayes scores in 3.0 and had a couple of questions:
> [...]
> 
> 
> ... the FAQ ... read the FAQ ...
> 

Great Bob, the FAQ says how the scores are generated, I surmised that.
But these questions aren't in the FAQ:

Does a human review the scores generated by the statistics engine?

Doesn't it make sense to have more of a bell curve on the 2nd set of bayes
scores?

If not, why not?

The teeth seem seem to be taken out of BAYES_99 with it's low 1.9 score,
and most of my spam is triggering .99 to 1.  That to me seems like an
obvious oversight, and I'm just wondering what the thinking was to leave it
at 1.9 for the 3.0 release.

Bayes scores in SA 3.0

[Chip Paswater]

 Hey guys,
 
 I was looking at the Bayes scores in 3.0 and had a couple of questions:
 
 score BAYES_00 0 0 -1.665 -2.599
 score BAYES_05 0 0 -0.925 -0.413
 score BAYES_20 0 0 -0.730 -1.951
 score BAYES_40 0 0 -0.276 -1.096
 score BAYES_50 0 0 1.567 0.001
 score BAYES_60 0 0 3.515 0.372
 score BAYES_80 0 0 3.608 2.087
 score BAYES_95 0 0 3.514 2.063
 score BAYES_99 0 0 4.070 1.886
 
 I'm running the full load, bayes, network tests, AWL.. everything, so I'm
 using the second set of scores.  It seems odd to me that BAYES_99 scores
 lower than BAYES_95, and BAYES_40 scores lower than BAYES_05.
 
 I figured these scores were probably generated from a statistics engine
 that used a corpus of spam and ham.  Did a human approve these scores
 before they were put into SA?  If so, what was the thinking behind them?
 
 I'm tempted to manually edit the scores to show more of a bell curve, but
 if there is some method here that I'm overlooking, I'd love to know what it
 is.