LISTSERV at the University of Georgia
Menubar Imagemap
Home Browse Manage Request Manuals Register
Previous messageNext messagePrevious in topicNext in topicPrevious by same authorNext by same authorPrevious page (March 2008, week 5)Back to main SAS-L pageJoin or leave SAS-L (or change settings)ReplyPost a new messageSearchProportional fontNon-proportional font
Date:         Sun, 30 Mar 2008 15:29:37 -0500
Reply-To:     Wensui Liu <liuwensui@GMAIL.COM>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         Wensui Liu <liuwensui@GMAIL.COM>
Subject:      Re: What exactly is DATA Mining?
Comments: To: Sigurd Hermansen <HERMANS1@westat.com>
In-Reply-To:  <CA8F89971ADA9F47A6C915BA2397844207B425FF@MAILBE2.westat.com>
Content-Type: text/plain; charset=ISO-8859-1

well, nice reading, sigurd, as a statistician, i have to admit that i DO hate some of socalled data miners from CS (computer science) area. data mining is so sweet and popular that they just want to share a spoon of honey out of it before they blow it up like DOTcom. however, look at how much fundamental research of data mining done by CS people? not even close to the contribution made by pioneer statisticians such as hastie, friedman, or tibshirani.

On Sun, Mar 30, 2008 at 3:04 PM, Sigurd Hermansen <HERMANS1@westat.com> wrote: > As an antidote to a statistician's *normal* perspective, see these rough > notes on "Why Statisticians Hate Us" [data miners] ... > http://www.cs.csi.cuny.edu/~imberman/DataMining/Statistics%20vs.pdf > > Many statisticians have come around to the notion that exploratory data > analysis complements inference. Data mining methods include linkage and > integration of data sources, data Q/C, visualization, and predictive > modelling. What we once call "data prep" for statistical analysis has > evolved in the same manner as statistical computation. Contemporary > research networks and longitudinal research databases rely on a > supporting platform of automated data mining methods. Even while > statisticians find it difficult to give up direct control of data > collection processes, they are finding new opportunities to expand the > scope of statistical research to rare events and complex patterns, > S > > > > -----Original Message----- > From: owner-sas-l@listserv.uga.edu [mailto:owner-sas-l@listserv.uga.edu] > On Behalf Of Peter Flom > Sent: Sunday, March 30, 2008 7:37 AM > To: Paula Sims; SAS-L@LISTSERV.UGA.EDU > Subject: Re: What exactly is DATA Mining? > > > Paula Sims <paulasims2004@HOTMAIL.COM> wroteeable ones. > > > >One of the big buzz topics during this year's SGF was/is Data Mining? > >What exactly is it? Is it just running some typical PROCS to check for > >structure and then regressions/correlations for interdependence? Is > >there anything you can suggest that I read up on the topic? My > >management is pretty clueless but they do get excited when new buzz > >words are introduced. > > Data mining consists of throwing huge piles of stuff at a wall, seeing > what sticks, and proclaiming it good :-) > > grin, duck, run > > Peter > > Statistical Consultant > www DOT peterflom DOT com >

-- =============================== WenSui Liu ChoicePoint Precision Marketing Phone: 678-893-9457 Email : wensui.liu@choicepoint.com Blog : statcompute.spaces.live.com ===============================


Back to: Top of message | Previous page | Main SAS-L page