Better initial configurations for metric multidimensional scaling

被引:11
作者
Malone, SW
Tarazaga, P
Trosset, MW
机构
[1] Duke Univ, Dept Math, Durham, NC 27708 USA
[2] Univ Puerto Rico, Dept Math, Mayaguez, PR 00681 USA
[3] Coll William & Mary, Dept Math, Williamsburg, VA 23185 USA
关键词
distance matrices; distance geometry; spectral decomposition; low rank approximations; stress and sstress criteria; global optimization;
D O I
10.1016/S0167-9473(02)00145-7
中图分类号
TP39 [计算机的应用];
学科分类号
081203 [计算机应用技术]; 0835 [软件工程];
摘要
Multidimensional scaling (MDS) is a collection of data analytic techniques for constructing configurations of points from dissimilarity information about interpoint distances. Two popular measures of how well the constructed distances fit the observed dissimilarities are the raw stress and sstress criteria, each of which must be minimized by numerical optimization. Because iterative procedures for numerical optimization typically find local minimizers that may not be global minimizers, the choice of an initial configuration from which to begin searching for an optimal configuration is crucial. A popular choice of initial configuration is the classical solution of Torgerson (Psychometrika 17 (1952) 401). Results from the theory of distance matrices are exploited to derive two alternatives, each guaranteed to be at least as good as the classical solution, and empirical evidence is presented that they are usually substantially better. (C) 2002 Elsevier Science B.V. All rights reserved.
引用
收藏
页码:143 / 156
页数:14
相关论文
共 29 条
[1]
[Anonymous], 1971, ATTITUDE RES REACHES
[2]
[Anonymous], 1952, Psychometrika
[4]
de Leeuw Jan., 1982, Handbook of statistics, V2, P285
[5]
DIMENSIONS OF COLOR VISION [J].
Ekman, Gosta .
JOURNAL OF PSYCHOLOGY, 1954, 38 (02) :467-474
[6]
THE EMBEDDING PROBLEM FOR PREDISTANCE MATRICES [J].
GLUNT, W ;
HAYDEN, TL ;
LIU, WM .
BULLETIN OF MATHEMATICAL BIOLOGY, 1991, 53 (05) :769-796
[7]
GOWER JC, 1966, BIOMETRIKA, V53, P325, DOI 10.2307/2333639
[8]
Green P. E., 1989, MULTIDIMENSIONAL SCA
[9]
Groenen PJ., 1993, MAJORIZATION APPROAC
[10]
The tunneling method for global optimization in multidimensional scaling [J].
Groenen, PJF ;
Heiser, WJ .
PSYCHOMETRIKA, 1996, 61 (03) :529-550