cats& dogs [was: New new script to train bogofilter]

[David Relson]

At 07:06 AM 7/4/03, Greg Louis wrote:

>So the problem you may see down the road if you train "to extinction"
>on errors is that bogofilter will get very good at recognizing the
>types of messages on which you train, and rather poor at recognizing
>messages that are similar to, but not strongly similar to, those
>training ones.  It's like trying to recognize dogs by training on
>German shepherds only: a great Dane shares the "dog" characteristics,
>but we'll have learned too many specific "German shepherd" ones so we
>may well misclassify the great Dane.

Assume we're using bogofilter to distinguish cats from dogs.  With the 
German shepherd training, bogofilter would likely think a chihuahua is a 
cat.  The size is clearly wrong for a German shepherd and is right for a cat.