| | |
Summary: DISTATIS: The Analysis of Multiple Distance Matrices
Herv´e Abdi
The University of Texas at Dallas
Dominique Valentin
Universit´e de Bourgogne
Alice J. O'Toole
The University of Texas at Dallas
Betty Edelman
The University of Texas at Dallas
Abstract
In this paper we present a generalization of classical
multidimensional scaling called DISTATIS which is a new
method that can be used to compare algorithms when their
outputs consist of distance matrices computed on the same
set of objects. The method first evaluates the similarity be-
tween algorithms using a coefficient called the RV coeffi-
cient. From this analysis, a compromise matrix is computed
which represents the best aggregate of the original matrices.
In order to evaluate the differences between algorithms, the
original distance matrices are then projected onto the com-
|
