The Structural Representation of Proximity Matrices with MATLAB
Buy a book... In Association with Amazon.co.uk
Author(s): L. Hubert, P. Arabie and J. Meulman
Publisher: SIAM
ISBN: 0898716071
Format: softback
214pp
Price: $79.00
Review Date: 12 April 2006
Review: This book explains how to actually use some 65 functions (by way of M-files) within a MATLAB computational environment to effect a variety of structural representations for proximity information assumed available on a set of objects. The structural representations have mainly been developed in the applied (behavioural sciences) statistical literature (in psychometrics and classification, for example). The book is divided into three main sections: Part I develops linear and circular uni- and multi- dimensional scaling using the city-block metric as the main representational device. Part II discusses the characterization of various graph-theoretic tree structures, in particular those usually referred to as ultrametrics and additive trees. The last part uses representations that are defined solely by order properties, particularly to what are called (strongly) anti-Robinson forms.