Re: [DT] Re: DM: binary vs. multiway splits in classification trees

[Tjen-Sien Lim]

   Mailing-List: contact decision_trees-owner@egroups.com
   X-Mailing-List: decision_trees@egroups.com
   X-URL: http://www.egroups.com/list/decision_trees/
   Reply-To: decision_trees@egroups.com
   Date: Fri, 26 Mar 1999 13:33:28 -0800 (PST)
   From: Dan Steinberg <dstein@salford-systems.com>
   cc: datamine-l@nautilus-sys.com, decision_trees@egroups.com
   Content-Type: text/plain; charset=US-ASCII

   An important question for FINITE data sets is which method works 
better in a
   specific context.  The multi-way split is a form of constraint and 
there will be
   occasions in which a constrained tree (multi-way) will outperform 
an
   unconstrained one (binary). However there will also be many 
circumstances in
   which the constrained tree is worse.


That's exactly what I'm after. I've several data sets (that I really
understand) where binary splits are better. I'm still looking for real
data sets that show multiway splits are preferred. I'll be grateful if
someone is willing to share his/her data. Thanks.

-- 
Tjen-Sien Lim                (608) 262-8181                        
Dept. of Statistics          limt@stat.wisc.edu
Univ. of Wisconsin-Madison   http://www.stat.wisc.edu/~limt
1210 West Dayton Street       
Madison, WI 53706