Tom Khabaza's 9 Laws

[Christophe Giraud-Carrier]

Tom Khabaza, who is one of the founders of ISL and developer of
Clementine (which was bought by SPSS and later by IBM) and a partner
in the early METAL project, recently came up with what he calls the "9
Laws of Data Mining" (see http://www.khabaza.com/9laws). I suspect
most of you will find these interesting, in particular the 4th Law,
which has to with the No Free Lunch and has thus direct implications
on research in Metalearning.

Tom recently told me that he felt that his "4th law of data mining,...
pretty much contradicts the ideas behind METAL." I have begun to think
a bit about that and would be interested about any comments or
thoughts anyone has on this.

Enjoy the reading. I look forward to your comments. Thanks,
Christophe