Exploring Topology Preservation of SOMs with a Graph Based Visualization

  • Kadim Taşdemir
Part of the Lecture Notes in Computer Science book series (LNCS, volume 5326)


The Self-Organizing Map (SOM), which projects a (high-dimensional) data manifold onto a lower-dimensional (usually 2-d) rigid lattice, is a commonly used manifold learning algorithm. However, a postprocessing – that is often done by interactive visualization schemes – is necessary to reveal the knowledge of the SOM. Thanks to the SOM property of producing (ideally) a topology preserving mapping, existing visualization schemes are often designed to show the similarities local to the lattice without considering the data topology. This can produce inadequate tools to investigate the detailed data structure and to what extent the topology is preserved during the SOM learning. A recent graph based SOM visualization, CONNvis [1], which exploits the underutilized knowledge of data topology, can be a suitable tool for such investigation. This paper discusses that CONNvis can represent the data topology on the SOM lattice despite the rigid grid structure, and hence can show the topology preservation of the SOM and the extent of topology violations.


Data Space Delaunay Triangulation Locally Linear Embedding Data Topology Nonlinear Dimensionality Reduction 
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.
    Taşdemir, K., Merényi, E.: Data topology visualization for the Self-Organizing Maps. In: Proc. 14th European Symposium on Artificial Neural Networks (ESANN 2006), Bruges, Belgium, D-Facto, April 26-28, pp. 277–282 (2006)Google Scholar
  2. 2.
    Cox, T.F., Cox, M.: Multidimensional Scaling. Chapman and Hall/CRC, Boca Raton (2001)zbMATHGoogle Scholar
  3. 3.
    Tenenbaum, J.B., de Silva, V., Langford, J.: A global geometric framework for nonlinear dimensionality reduction. Science 290(5500), 2319–2323 (2000)CrossRefGoogle Scholar
  4. 4.
    Roweis, S., Soul, L.: Nonlinear dimensionality reduction by locally linear embedding. Science 290(5500), 2323–2326 (2000)CrossRefGoogle Scholar
  5. 5.
    Kohonen, T.: Self-Organizing Maps, 2nd edn. Springer, Heidelberg (1997)CrossRefzbMATHGoogle Scholar
  6. 6.
    Yin, H.: Nonlinear multidimensional data projection and visualisation. In: Liu, J., Cheung, Y.-m., Yin, H. (eds.) IDEAL 2003. LNCS, vol. 2690, pp. 377–388. Springer, Heidelberg (2003)Google Scholar
  7. 7.
    Ultsch, A.: Self-organizing neural networks for visualization and classification. In: Lausen, O.B., Klar, R. (eds.) Information and Classification-Concepts, Methods and Applications, pp. 307–313. Springer, Berlin (1993)CrossRefGoogle Scholar
  8. 8.
    Kraaijveld, M., Mao, J., Jain, A.: A nonlinear projection method based on Kohonen’s topology preserving maps. IEEE Trans. on Neural Networks 6(3), 548–559 (1995)CrossRefGoogle Scholar
  9. 9.
    Cottrell, M., de Bodt, E.: A Kohonen map representation to avoid misleading interpretations. In: Proc. 4th European Symposium on Artificial Neural Networks (ESANN 1996), Bruges, Belgium, D-Facto, pp. 103–110 (1996)Google Scholar
  10. 10.
    Ultsch, A.: Maps for the visualization of high-dimensional data spaces. In: Proc. 4th Workshop on Self-Organizing Maps (WSOM 2003), vol. 3, pp. 225–230 (2003)Google Scholar
  11. 11.
    Kaski, S., Kohonen, T., Venna, J.: Tips for SOM processing and colourcoding of maps. In: Deboeck, T.K.G. (ed.) Visual Explorations in Finance Using Self-Organizing Maps, London (1998)Google Scholar
  12. 12.
    Himberg, J.: A SOM based cluster visualization and its application for false colouring. In: Proc. IEEE-INNS-ENNS International Joint Conf. on Neural Networks, Como, Italy, vol. 3, pp. 587–592 (2000)Google Scholar
  13. 13.
    Vesanto, J.: SOM-based data visualization methods. Intelligent Data Analysis 3(2), 111–126 (1999)CrossRefzbMATHGoogle Scholar
  14. 14.
    Taşdemir, K., Merényi, E.: Exploiting data topology in visualization and clustering of Self-Organizing Maps. IEEE Transactions on Neural Networks (submitted)Google Scholar
  15. 15.
    Merkl, D., Rauber, A.: Alternative ways for cluster visualization in Self-Organizing Maps. In: Proc. 1st Workshop on Self-Organizing Maps (WSOM 2005), Espoo, Finland, Helsinki University of Technology, June 4-6, pp. 106–111. Helsinki University of Technology, Neural Networks Research Centre, Espoo, Finland (1997)Google Scholar
  16. 16.
    Su, M.-C., Chang, H.-T.: A new model of self-organizing neural networks and its applications. IEEE Transactions on Neural Networks 12(1), 153–158 (2001)CrossRefGoogle Scholar
  17. 17.
    Yin, H.: ViSOM- a novel method for multivariate data projection and structure visualization. IEEE Transactions on Neural Networks 13(1), 237–243 (2002)CrossRefGoogle Scholar
  18. 18.
    Yin, H.: Resolution enhancement for the ViSOM. In: Proc. 4th Workshop on Self-Organizing Maps (WSOM 2003), pp. 208–212 (2003)Google Scholar
  19. 19.
    Merényi, E., Taşdemir, K., Farrand, W.: Intelligent information extraction to aid science decision making in autonomous space exploration. In: Proc. SPIE Defense and Security, Orlando, Florida, March 17-20 (2008)Google Scholar
  20. 20.
    Martinetz, T., Schulten, K.: Topology representing networks. Neural Networks 7(3), 507–522 (1993)CrossRefGoogle Scholar
  21. 21.
    Ultsch, A.: Clustering with som: U*c. In: Proc. 5th Workshop on Self-Organizing Maps (WSOM 2005), Paris, France, September 5-8, 2005, pp. 75–82 (2005)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2008

Authors and Affiliations

  • Kadim Taşdemir
    • 1
  1. 1.Computer Engineering DYO KampusuYasar UniversityBornovaTurkey

Personalised recommendations