Rough Sets (Was: Berkeley Seminar)

[rbl@xxxxxxxxxxxxxxxxx (Robin B. Lake)]

I did some mining on the Web and came up with a presentation on
"Rough Sets".  When some of you e-mailed me and asked what they were,
I had to admit I didn't know.  So, look at:

http://www.trolldata.no/torvill/present/pkdd97/index.htm

Having gone thru this brief presentation, I still don't know what "rough
sets" are, but Mr. Torvill's thesis work says:

Abstract        This thesis attempts to deal with the problem of time within the
 framework of rough sets. The rough set theory has emphasized the reduction of i
nformation necessary to acquire desired knowledge.

==
As I understand it, rough sets are used to reduce the dimensionality and
complexity of data sets.  Examples include stock market data (time)
and maps (spatial patterns).

IMHO, there may be better-tested methods.

Cheers,
Rob Lake