Re: DM: Missing items in clustering!

[Yoram Reich]

In the context of hierarchical clustering, ECOBWEB (a program that 
extends
COBWEB in several dimensions including handling continuous 
attributes) is
capable of handling missing attributes naturally. The category utility
function that is used for assessing the quality of the clustering at 
each
level of the hierarchy will simply have missing terms.  In the past, I
looked into some properties of category utility in relation to 
irrelevant
and other types of attributes (http://or.eng.tau.ac.il:7777/ai.ps.Z)  
but
did not completed an analysis of missing values. I am sure that such
analysis can be done. More about ECOBWEB can be found in: 
http://or.eng.tau.ac.il:7777/topics/ecobweb.html

cheers,
Yoram

On Tue, 7 Oct 1997, Rajan Kumaralingam wrote:

> Hi all,
> Does anyone have any pointers to how to handle missing 
> items in a clustering context (I'm currently using Euclidean metric
> based clustering).
> 
> Thanks in advance
> Raj
> 

Department of Solid Mechanics, Materials and Structures,
   Faculty of Engineering, Tel Aviv University, Ramat Aviv 69978, 
Israel
      Tel: + 972 3 6407385, Fax: + 972 3 6407617
         email: yoram@eng.tau.ac.il, http://or.eng.tau.ac.il:7777/