Visualization of High-Dimensional Data with Relational Perspective Map

Abstract

This paper introduces a method called relational perspective map (RPM) to visualize distance information in high-dimensional spaces. Like conventional multidimensional scaling, the RPM algorithm aims to produce proximity preserving 2-dimensional (2-D) maps. The main idea of the RPM algorithm is to simulate a multiparticle system on a closed surface: whereas the repulsive forces between the particles reflect the distance information, the closed surface holds the whole system in balance and prevents the resulting map from degeneracy. A special feature of RPM algorithm is its ability to partition a complex dataset into pieces and map them onto a 2-D space without overlapping. Compared to other multidimensional scaling methods, RPM is able to reveal more local details of complex datasets. This paper demonstrates the properties of RPM maps with four examples and provides extensive comparison to other multidimensional scaling methods, such as Sammon Mapping and Curvilinear Principle Analysis.

Keywords

multidimensional scaling dimensionality reduction

Get full access to this article

View all access options for this article.

References

Cox

Multidimensional Scaling. Heinemann: 1994.

Borg

Groenen

Modern Multidimensional Scaling. Spring: 1997.

Sammon

. A nonlinear mapping algorithm for data structure analysis. IEEE Transaction on Computers 1969; C18: 401–409.

Ramsay

. Maximum likelihood estimation in multidimensional scaling. Psychometricka 1977; 42: 241–266.

Demartines

Hérault

Curvilinear component analysis: A self-organizing neural network for nonlinear mapping of data sets. IEEE Transaction on Neural Networks 1997; 8: 148–154.

Tenenbaum

Silva de

Langford

. A global geometric framework for nonlinear dimensionality reduction. Science 2000; 290: 2319–2323.

Lee

Lendasse

Donckers

Verleysen

A robust nonlinear projection method. In: Proceedings of ESANN' 2000, 8th European Symposition on Artificial Neural Networks (Bruges, Belgium, 2000), M. D-Facto public, 13–20.

Lee

Lendasse

Verleysen

Curvlinear distance analysis versus Isomap. In: Proceedings of European Symposium on Artifical Neural Networks (Bruges, Belgium, 2002); d-side publications, 185–192.

Morrison

Ross

Chalmers

Fast multidimensional scaling, springs and interpolation. Information Visualization 2003; 2: 68–77.

10.

Kohonen

Self-organized formation of topologically correct feature maps. Biological Cybernetics 1982; 43: 59–69.

11.

Fruchterman

Rheingold

. Graph drawing by force-directed placement. Software – Practice and Experience 1991; 21: 1129–1164.

12.

Ashby

Brittin

. Thomson's problem. American Journal of Physics 1986; 54: 776–777.

13.

van Laarhoven

PJM

Aarts

EHL

. Simulated Annealing: Theory and Application. Dordrect: Reidel, 1987.

14.

Press

Teukolsky

Vetterling

Flannery

. Numerical Recipes in C++ (2nd Edition). Cambridge University Press: Cambridge, 2002, 366–372.

15.

VisuMap Technologies, VisuMap version 1.4 (standard edition). The software program used to produce examples in this paper. This program and associated datasets can be downloaded at http://www.visumap.net (accessed August 9, 2003).

16.

Hérault

Guérin-Dugué

Villemain

Searching for the embedded manifolds in high-dimensional data, problems and unsolved questions. European Conference on Artificial Neural Networks (Bruges, Belgium, 2002).