Pattern Classification Using a Set of Compact Hyperspheres

  • Amir Atiya
  • Sherif Hashem
  • Hatem Fayed
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 4233)


Prototype classifiers are one of the simplest and most intuitive approaches in pattern classification. However, they need careful positioning of prototypes to capture the distribution of each class region. Classical methods, such as learning vector quantization (LVQ), are sensitive to the initial choice of the number and the locations of the prototypes. To alleviate this problem, a new method is proposed that represents each class region by a set of compact hyperspheres. The number of hyperspheres and their locations are determined by setting up the problem as a set of quadratic optimization problems. Experimental results show that the proposed approach significantly beats LVQ and Restricted Coulomb Energy (RCE) in most performance aspects.


Pattern Classification Learn Vector Quantization Quadratic Optimization Problem Class Region Prototype Classifier 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Hastie, T., Tibshirani, R.: Discriminant adaptive nearest neighbor classification. IEEE Trans. Pattern Analysis and Machine Intelligence. 18(6), 607–616 (1996)CrossRefGoogle Scholar
  2. 2.
    Reilly, D., Cooper, L., Elbaum, C.: A neural model for category learning. Biological Cybernetics 45, 35–41 (1982)CrossRefGoogle Scholar
  3. 3.
    Hudak, M.J.: RCE Classifiers: Theory and Practice. Cybernetics and Systems: An International Journal 23, 483–515 (1992)CrossRefMathSciNetGoogle Scholar
  4. 4.
    Tsumura, N., Itoh, K., Ichioka, Y.: Reliable classification by double hyperspheres in pattern vector space. Pattern Recognition 28, 1621–1626 (1995)CrossRefGoogle Scholar
  5. 5.
    Kositsky, M., Ullman, S.: Learning Class Regions by the Union of Ellipsoids. In: Proc. 13th International Conference on Pattern Recognition (ICPR), vol. 4, pp. 750–757. IEEE Computer Society Press, Los Alamitos (1996)CrossRefGoogle Scholar
  6. 6.
    Atiya, A., Hashem, S., Fayed, H.: New hyperspheres for pattern classification. In: Proc. 1st International Computer Engineering Conference (ICENCO-2004), Cairo, Egypt, pp. 258–263 (2004)Google Scholar
  7. 7.
    Wright, S.J.: Primal-Dual Interior-Point Methods. SIAM Publications, Philadelphia (1997)MATHGoogle Scholar
  8. 8.
    Khachiyan, L.G.: Rounding of polytopes in the real number model of computation. Mathematics of Operations Research 21, 307–320 (1996)MATHCrossRefMathSciNetGoogle Scholar
  9. 9.
    Kohonen, T.: Self-organization and Associative Memory, 3rd edn. Springer, Heidelberg (1989)Google Scholar
  10. 10.
    Hastie, T., Tisbshirani, R., Friedman, J.: The Elements of Statistical Learning. Springer, Heidelberg (2001)MATHGoogle Scholar
  11. 11.
    Zhang, H., Sun, G.: Optimal reference subset selection for nearest neighbor classification by tabu search. Pattern Recognition 35, 1481–1490 (2002)MATHCrossRefGoogle Scholar
  12. 12.
    Prechelt, L.: Proben1. A Set of Neural-Network Benchmark Problems. Univ. Karlsruhe, Germany (1994), available FTP:
  13. 13.
    Michie, D., Spiegelhalter, D.J., Taylor, C.C.: Machine Learning, Neural and Statistical Classification. Ellis Horwood, New York (1994)MATHGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2006

Authors and Affiliations

  • Amir Atiya
    • 1
  • Sherif Hashem
    • 2
  • Hatem Fayed
    • 2
  1. 1.Computer Engineering DepartmentCairo UniversityEgypt
  2. 2.Engineering Mathematics and Physics DepartmentCairo UniversityEgypt

Personalised recommendations