Convergence Rate in Intelligent Self-organizing Feature Map Using Dynamic Gaussian Function

  • Geuk Lee
  • Seoksoo Kim
  • Tai Hoon Kim
  • Min Wook Kil
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 4252)


The existing self-organizing feature map has weak points when it trains. It needs too many input patterns, and a learning time is increased to handle them. In this paper, we propose a method improving the convergence speed and the convergence rate of the intelligent self-organizing feature map by adapting Dynamic Gaussian Function instead of using a Neighbor Interaction Set whose learning rate is steady during the training of the self-organizing feature map.


Neural Network Convergence Rate Input Vector Learning Rate Convergence Speed 
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.
    Kohonen, T.: The Neural Phonetic Typewriter. computer 21(3), 11–22 (1988)CrossRefGoogle Scholar
  2. 2.
    Kohonen, T.: Self-Organization and Associative Memory. Springer, Heidelberg (1989)Google Scholar
  3. 3.
    Kohonen, T.: The Self-Organizing Map. In: Proc. of the IEEE, vol. 78, pp. 1464–1480 (September 1990)Google Scholar
  4. 4.
    Paolo, L., Patrick, T., Nikolaos, V.: Modified self-organizing feature map algorithms for efficient digital hardware implementation. IEEE trans. on neural network 8, 315–330 (1997)CrossRefGoogle Scholar
  5. 5.
    George, B.: Georgiopoulous, Michael, Using self-organizing maps to learn geometric hash function for model-based recognition. IEEE trans. on neural network 9, 560–570 (1998)CrossRefGoogle Scholar
  6. 6.
    Teubo Kohonen, E., Oja, A., Visa, O., Simula, O.: Enginnering applications of self-organizing map. In: Proc. of the IEEE, pp. 1358–1384 (1996)Google Scholar
  7. 7.
    Demartines, P., Herault, J.: Curvilinear component analysis: a self-organizing neural network for nonlinear mapping of data sets. IEEE trans. on neural network 8, 148–154 (1997)CrossRefGoogle Scholar
  8. 8.
    Kohonen, T.: Generalizations of the Self-Organizing Map. In: Proc. of IJCNN, vol. 1, pp. 457–461 (1993)Google Scholar
  9. 9.
    Lo, Z.P., Bavarian, B.: Two Theorem for the Kohonen Mapping Neural Network. In: Proc. of IJCNN, vol. 4, pp. 755–760 (1992)Google Scholar
  10. 10.
    Lo, Z.P., Yu, Y., Bavarian, B.: Convergence Properties of Topology Preserving Neural Network. IEEE Trans. Neural Networks 4(2) (March 1993)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2006

Authors and Affiliations

  • Geuk Lee
    • 1
  • Seoksoo Kim
    • 2
  • Tai Hoon Kim
    • 3
  • Min Wook Kil
    • 4
  1. 1.Department of Computer Eng.Hannam UniversityDaeJeonSouth Korea
  2. 2.Department of Multimedia.Hannam UniversityDaeJeonSouth Korea
  3. 3.Security Engineering Research GroupDaeJeonSouth Korea
  4. 4.Dept. of Medical Inform.Mun Kyung CollegeHoGyeMyun, Mun KyungSouth Korea

Personalised recommendations