Print

Print


First of all, MDS is NOT a clustering method, although the two methodologies are often used in a complementary manner, just as, say, principal components and clustering analysis may sometimes be (but often in an inappropriate way).

Secondly, there's an enormous literature on both subjects, including numerous papers discussing the two approaches and their interrelationships.  Two papers that were recently completed that might be helpful to you, both to be published in the forthcoming Encyclopedia of Social Measurement, are one by Carroll, Arabie and Hubert on MDS and another by Arabie, Hubert and Carroll on Clustering.  These are both short overview articles, but provide other references which could be useful to you.  Since these papers are not yet published, if you're interested I could arrange to send you electronic copies of preprints of both.

Doug Carroll

At 11:13 AM 5/5/2004 -0500, Henry Bulley wrote:

Hello all,

 

Do any of you know of a good review paper/article that discusses cluster analysis/cluster validation as well as other clustering approaches like MDS?

 

Thanks

 

Henry

 

 

From: Classification, clustering, and phylogeny estimation [mailto:[log in to unmask]] On Behalf Of F. James Rohlf
Sent: Monday, March 22, 2004 8:23 PM
To: [log in to unmask]
Subject: Re: Question about cluster analysis techniques

 

You might find more responses to such questions on the morphometrics list server. For information about joining this list see http://life.bio.sunysb.edu/morph/morphmet.html. You might also find this website useful has it has links to many resources for the study of shape variation.

-----------------------
F. James Rohlf
State University of New York, Stony Brook, NY 11794-5245
www: http://life.bio.sunysb.edu/ee/rohlf
-----Original Message-----
From: Classification, clustering, and phylogeny estimation [mailto:[log in to unmask]] On Behalf Of Angelique Augereau
Sent: Wednesday, March 17, 2004 6:23 PM
To: [log in to unmask]
Subject: Question about cluster analysis techniques
Hi,
 
I have human body scan data on 4 parts of the body: waist, upper hip, high hip and lower hip.  I am trying to find a way to classify human bodies into different types based on the 4 measures.  E.g. straight body, pear shape, etc.
 
Can you please tell me what kinds of techniques would be appropriate for this kind of analysis?  I am working with Mathematica.
 
Thanks!



  ######################################################################
  # J. Douglas Carroll, Board of Governors Professor of Management and #
  #Psychology, Rutgers University, Graduate School of Management,      #
  #Marketing Dept., MEC125, 111 Washington Street, Newark, New Jersey  #
  #07102-3027.  Tel.: (973) 353-5814, Fax: (973) 353-5376.             #   
  # Home: 14 Forest Drive, Warren, New Jersey 07059-5802.              #   
  # Home Phone: (908) 753-6441 or 753-1620, Home Fax: (908) 757-1086.  #   
  # E-mail: [log in to unmask]                                   #
  ######################################################################