Cluster Validity Using Support Vector Machines

  • Vladimir Estivill-Castro
  • Jianhua Yang
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 2737)


Gaining confidence that a clustering algorithm has produced meaningful results and not an accident of its usually heuristic optimization is central to data analysis. This is the issue of validity and we propose here a method by which Support Vector Machines are used to evaluate the separation in the clustering results. However, we not only obtain a method to compare clustering results from different algorithms or different runs of the same algorithm, but we can also filter noise and outliers. Thus, for a fixed data set we can identify what is the most robust and potentially meaningful clustering result. A set of experiments illustrates the steps of our approach.


Support Vector Machine Support Vector Cluster Result Cluster Structure Potential Outlier 
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.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Bennett, K.P., Campbell, C.: Support vector machines: Hype or hallelujah. SIGKDD Explorations 2(2), 1–13 (2000)CrossRefGoogle Scholar
  2. 2.
    Bezdek, J.C.: Pattern Recognition with Fuzzy Objective Function Algorithms. Plenum, New York (1981)zbMATHGoogle Scholar
  3. 3.
    Chang, C.C., Lin, C.J.: LIBSVM: a library for support vector machines (2001),
  4. 4.
    Chang, C.C., Lin, C.J.: Training v-support vector classifiers: Theory and algorithms. Neural Computation 13(9), 2119–2147 (2001)zbMATHCrossRefGoogle Scholar
  5. 5.
    Cherkassky, V., Muller, F.: Learning from Data — Concept, Theory and Methods. Wiley, New York (1998)Google Scholar
  6. 6.
    Dubes, R.C.: Cluster analysis and related issues. In: Chen, C.H., Pau, L.F., Wang, P.S.P. (eds.) Handbook of Pattern Recognition and Computer Vision, ch. 1.1, pp. 3–32. World Scientific, Singapore (1993)Google Scholar
  7. 7.
    Estivill-Castro, V.: Why so many clustering algorithms - a position paper. SIGKDD Explorations 4(1), 65–75 (2002)CrossRefMathSciNetGoogle Scholar
  8. 8.
    Gokcay, E., Principe, J.: A new clustering evaluation function using Renyi’s information potential. In: Wells, R.O., Tian, J., Baraniuk, R.G., Tan, D.M., Wu, H.R. (eds.) Proc. of IEEE Int. Conf. Acoustics, Speech and Signal Processing (ICASSP 2000), Istanbul, pp. 3490–3493 (2000)Google Scholar
  9. 9.
    Gunn, S.: Support vector machines for classification and regression. Tech. Report ISIS-1-98, Univ. of Southampton, Dept. of Electronics and Computer Science (1998)Google Scholar
  10. 10.
    Haykin, S.S.: Neural networks: a comprehensive foundation. PrenticeHall, Englewood Cliffs (1999)zbMATHGoogle Scholar
  11. 11.
    Jain, A.K., Dubes, R.C.: Algorithms for Clustering Data. PrenticeHall, Englewood Cliffs (1998)Google Scholar
  12. 12.
    Koschke, R., Eisenbarth, T.: A framework for experimental evaluation of clustering techniques. In: Proc. Int. Workshop on Program Comprehension (2000)Google Scholar
  13. 13.
    Rauber, A., Paralic, J., Pampalk, E.: Empirical evaluation of clustering algorithms. Malekovic, M., Lorencic, A. (eds.), 11th Int. Conf. Information and Intelligent Systems (IIS 2002), Varazdin, Croatia, September 20 - 22, Univ. of Zagreb (2000)Google Scholar
  14. 14.
    Schölkopf, B., Williamson, R.C., Smola, A.J., Shawe-Taylor, J.: SV estimation of a distribution’s support. In: Leen, T.K., Solla, S.A., Müller, K.R. (eds.) Advances in Neural Information Processing Systems 12. MIT Press, Cambridge (forthcomming),
  15. 15.
    Siegelmann, H., Ben-Hur, A., Horn, D., Vapnik, V.: Support vector clustering. J. Machine Learning Research 2, 125–137 (2001)Google Scholar
  16. 16.
    Vapnik, V.N.: The nature of statistical learning theory. Springer, Heidelberg (1995)zbMATHGoogle Scholar
  17. 17.
    Vazirgiannis, M., Halkidi, M., Batistakis, Y.: On clustering validation techniques. Intelligent Information Systems J. 17(2), 107–145 (2001)zbMATHCrossRefGoogle Scholar
  18. 18.
    Williamson, R., Schölkopf, B., Smola, A., Bartlett, P.: New support vector algorithms. Neural Computation 12(5), 1207–1245 (2000)CrossRefGoogle Scholar
  19. 19.
    Winter, R.: Formal validation of schema clustering for large information systems. In: Proc. First American Conference on Information Systems (1995)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2003

Authors and Affiliations

  • Vladimir Estivill-Castro
    • 1
  • Jianhua Yang
    • 2
  1. 1.Griffith UniversityBrisbaneAustralia
  2. 2.The University of Western SydneyCampbelltownAustralia

Personalised recommendations