"T.S. Lim" <Use-Author-Address-Header@[127.1]> wrote: > In the Data Mining world that is dominated by Computer Scientists, the > methodology behind the software packages sold/licensed in the market > is often proprietary. Take, for example, the classification and > regression trees software package CART(r). The basic idea behind > CART(r) is the algorithm proposed by Breiman, Friedman, Olshen, and > Stone (1984). However, there has been quite a few proprietary > improvement in CART(r) so that you can no longer know for sure what's > going on inside the software package. The same is true for C5.0/See5 > (another classification trees software) that supersedes C4.5. > > When dealing with proprietary methodology, it's (practically) > impossible to study the properties of the method > thoroughly. Personally, I feel uncomfortable using a method that can't > be evaluated objectively by fellow researchers. Agreed. I wouldn't use it. It may be OK if the > application has nothing to do with human experimentation (as in > Biostatistics). > Since most (if not all) applications of Data Mining > are in commerce, the risk of using unproven methodology that hasn't > been extensively scrutinized may be acceptable. Most of the stuff being done are hyped-up hacks. They wouldn't dare publish this junk lest someone with some knowledge tear it apart. It is fairly easy to impress banking execs with the right buzz words and marketing spin. The mathematical/statistical validity of the technology is really secondary at best. Each co. will have their favorite tool/approach and will try to fit every problem into the framework of this tool/approach - sometimes without much thought. > Perhaps this joke is true after all: when a Statistician gets an idea, > she/he'll write and publish a paper while when a Computer Scientist > gets an idea, she/he'll form a company. :) > It sucks but this has been true in my experience. > Comments? > > -- > T.S. Lim > [EMAIL PROTECTED] > www.Recursive-Partitioning.com > > ------------------------------------------------------------ > Get paid to write review! http://recursive-partitioning.epinions.com > > Sent via Deja.com http://www.deja.com/ Before you buy. ================================================================= Instructions for joining and leaving this list and remarks about the problem of INAPPROPRIATE MESSAGES are available at http://jse.stat.ncsu.edu/ =================================================================
