RE: DM: Bagging, Boosting, Bundling capabilities

 >Date: Sat, 27 May 2000 09:38:42 -0700 (PDT)
 >From: Dan Steinberg <dstein@salford-systems.com>
 >To: "'datamine-l@nautilus-sys.com'" <datamine-l@nautilus-sys.com>
 >Subject: DM: Bagging, Boosting, Bundling capabilities
 >Reply-To: datamine-l@nautilus-sys.com
 >
 >
 >In a recent exchange Tom Khabaza quoted "Ken":
 >
 >  > bagging, and boosting to improve our predictive accuracy.
 >  > However, as far as
 >  > I know SAS EM is the only tool that has built in this
 >  > capability.
 >
 >Salford Systems CART(r) has bagging and the ARCing variant of boosting built
 >in; bundling wil follow shortly.  We've been shipping UNIX based versions of
 >CART with bagging since 1996.
 >
 >
 >  *---------------------------+---------------------------------*
 >  | Dan Steinberg             | FAX (619) 543 8888              |
 >  | Salford Systems           | VOICE (619) 543-8880            |
 >  | 8880 Rio San Diego Dr     |                                 |
 >  | San Diego, CA 92108       | http://www.salford-systems.com  |
 >  *-------------------------------------------------------------*


C5.0 also boosting capability. Boosting/bagging/arcing indeed can improve 
the prediction accuracies of classification trees. However, they can fail 
sometimes. Visit

    http://www.recursive-partitioning.com/mv.shtml

to see my empirical comparisons.



--
T.S. Lim
tslim@recursive-partitioning.com
www.Recursive-Partitioning.com



------------------------------------------------------------
Get paid to write review! http://recursive-partitioning.epinions.com

• Prev by Date: Re: DM: AW: Classification problem
• Next by Date: Re: AW: DM: RE: Data Forms for Mining (Limit on variables)
• Prev by thread: DM: Bagging, Boosting, Bundling capabilities
• Next by thread: DM: information retrieval
• Index(es):
  • Date
  • Thread