RE: [S] Data mining

Gunter, Bert (
Mon, 11 May 1998 08:18:37 -0400

I felt it necessary to reply briefly to the following, just in case Prof.
Ripley is too modest (!!?) to do so. The terms data mining/KDD have by now
been so flaunted and abused that they no longer mean much of anything.
However, if they were to be meaningful, one would assume that there ought to
be a large intersection with data analysis (and design ?) procedures for
large data sets. For which, IMHO, the best and most comprehensive reference
is Brian Ripley's book, PATTERN RECOGNITION AND NEURAL NETS, information for
which may be found at his web site, .
In particular, I would say that this presents a body of knowledge that is
too often ignored by KDD'ers, especially those in the CS community, who,
alas, often seem more intent on creating algorithms than in understanding
the nature of the problems that they deal with and finding out about the
methodology that may already be available to deal with them. 'Nuff said;
dissenting opinions invited.

> During my research, I found some white papers which has Data Mining
> concepts and some methods but it is not good enough to apply to a real
> case.
> If you want to have a look at what I found.
> -Book : advances in knowledge discovery and data mining.
>         Fayyad, Piatetsky-Shapiro, Smyth , Uthurusamy
> -web site :
>             h ttp://
> These are the most relarvent materials.
> If you want more then let me know.
> Thank you very much.
> Yours Sincerely,
> So Jung Lee
Disclaimer: The views expressed here are mine and not those of Merck or my
colleagues at Merck.

Bert Gunter
Biometrics Research
Merck Research Labs
P.O. Box 2000
Rahway, NJ 07065-0900

"The business of the statistician is to catalyze the
scientific learning process." George E. P. Box

This message was distributed by To unsubscribe
send e-mail to with the BODY of the
message: unsubscribe s-news