Advertisement

On the Interface between Cluster Analysis, Principal Component Analysis, and Multidimensional Scaling

  • H. H. Bock
Part of the Theory and Decision Library book series (TDLB, volume 8)

Abstract

This paper shows how methods of cluster analysis, principal component analysis, and multidimensional scaling may be combined in order to obtain an optimal fit between a classification underlying some set of objects 1,…,n and its visual representation in a low-dimensional euclidean space ℝs. We propose several clustering criteria and corresponding k-means-like algorithms which are based either on a probabilistic model or on geometrical considerations leading to matrix approximation problems. In particular, a MDS-clustering strategy is presented for-displaying not only the n objects using their pairwise dissimilarities, but also the detected clusters and their average distances.

Keywords

Multidimensional Scaling Dissimilarity Matrix Iterative Relaxation Basic Principal Component Analysis Matrix Approximation Problem 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. Bezdek, J.C., Ch. Coray, R. Gunderson, J. Watson: Detection and characterization of cluster substructure. I. Linear structure: Fuzzy c-lines. II, Fuzzy c-varietes and convex combinations thereof. SIAM J. Appl. Math. 40 (1981) 339–357MathSciNetzbMATHCrossRefGoogle Scholar
  2. Bezdek, J.C., Ch. Coray, R. Gunderson, J. Watson: Detection and characterization of cluster substructure. I. Linear structure: Fuzzy c-lines. II, Fuzzy c-varietes and convex combinations thereof. SIAM J. Appl. Math. 40 (1981) 358–372.MathSciNetzbMATHCrossRefGoogle Scholar
  3. Bock, H.H.: Automatische, Klassifikation. Theoretische und praktische Methoden zur, Gruppierung und Strukturierung von Daten (Cluster-Analyse). Vandenhoeck & Ruprecht, Göttingen 1974, 480 pp.zbMATHGoogle Scholar
  4. Bock, H.H.: Clusteranalyse mit unscharfen Partitionen. In: Bock, H.H. (ed.): Klassifikation und Erkenntnis III: Numerische Klassifikation. Studien zur Klassifikation SK-6. INDEKS-Verlag, Frankfurt 1979, 137–163.Google Scholar
  5. Bock, H.H.: Multidimensional scaling in the framework of cluster analysis. In: Degens, P.O., H.-J. Hermes, O. Opitz (eds.): Classification and its environment. (Proc. 10. Annual Meeting of the Gesellschaft für Klassifikation) Studien zur Klassifikation SK-17. INDEKS-Verlag, Frankfurt 1986.Google Scholar
  6. Diday, E. et al.: Optimisation en classification automatique I, II. Institut National de Recherche en Informatique et en Automatique (INRIA), Le Chesnay/France 1979.Google Scholar
  7. Friedman, H.P., J. Rubin: On some invariant criteria for grouping data. J. Amer. Statist. Assoc. 62 (1967) 1159–1178.MathSciNetCrossRefGoogle Scholar
  8. Huber, P.J.: Projection pursuit. Annals of Statistics 13 (1985) 435–475.MathSciNetzbMATHCrossRefGoogle Scholar
  9. Keller, J.B.: Factorization of matrices by least squares. Biometrika 49 (1962) 239–242.MathSciNetzbMATHGoogle Scholar
  10. Mathar, R.: The best Euclidean fit to a given distance matrix in prescribed dimensions. Linear Algebra and Its Applications 67 (1985) 1–6.MathSciNetzbMATHCrossRefGoogle Scholar
  11. Okamoto, M., M. Kanazawa: Minimisation of eigenvalues of a matrix and optimality of principal components. Ann. Math. Statist. 39 (1968) 859–863.MathSciNetzbMATHCrossRefGoogle Scholar
  12. Pearson, K.: On lines and planes of closest fit to systems of points in space. Phil. Mag. (6th series) 2 (1901) 559–572.Google Scholar
  13. Rao, C.R.: The use and interpretation of_principal component analysis in applied research. Sankkya A 26 (1964) 329–358.Google Scholar

Copyright information

© D. Reidel Publishing Company, Dordrecht, Holland 1987

Authors and Affiliations

  • H. H. Bock
    • 1
  1. 1.Institute of StatisticsTechnical University AachenAachenGermany

Personalised recommendations