Improving the Self-Organizing Feature Map Algorithm Using an Efficient Initialization Scheme

Mu-Chun Su; Ta-Kang Liu; Hsiao-Te Chang

doi:10.6180/jase.2002.5.1.05

Improving the Self-Organizing Feature Map Algorithm Using an Efficient Initialization Scheme

Computer Science and Information Engineering

Mu-Chun Su This email address is being protected from spambots. You need JavaScript enabled to view it.¹ , Ta-Kang Liu² and Hsiao-Te Chang²

¹Department of Computer Science and Information Engineering National Central University Chung Li, Taiwan 320, R.O.C.
²Department of Electrical Engineering Tamkang University Tamsui, Taipei, Taiwan 251, R.O.C.

Received: December 19, 2001
Accepted: February 18, 2002
Publication Date: March 1, 2002

Download Citation: ||https://doi.org/10.6180/jase.2002.5.1.05

ABSTRACT

It is often reported in the technique literature that the success of the self-organizing feature map formation is critically dependent on the initial weights and the selection of main parameters (i.e. the learning-rate parameter and the neighborhood set) of the algorithm. They usually have to be counteracted by the trial-and-error method; therefore, often time consuming retraining procedures have to precede before a neighborhood preserving feature amp is obtained. In this paper, we propose an efficient initialization scheme to construct an initial map. We then use the self-organizing feature map algorithm to make small subsequent adjustments so as to improve the accuracy of the initial map. Several data sets are tested to illustrate the performance of the proposed method.

Keywords: Neural Networks, Self-organizing Feature Map, Unsupervised Learning, Kohonen Algorithm

REFERENCES

[1] Baraldi, A., Blonda, P., Parmiggiani, F., Pasquariello, G. and Satalino, G., “Model Transitions in Descending FLVQ, ＂ IEEE Trans. on Neural Networks, Vol. 9, pp. 724-738 (1998).
[2] Fritzke, B., “Growing Cell Structures-a Self-Organizing Network for Unsupervised and Supervised Learning,” Neural Networks, Vol. 7, pp. 1441-1460 (1994).
[3] Harp, S. A. and Samad, T., “Genetic Optimization of Self-Organizing Feature Maps,” Proc. Int. Conf. on Neural Networks, pp. 341-346, (1991).
[4] Huang, S. J. and Hung, C. C., “Genetic Algorithms Enhanced Kohonen’s Neural Networks,” IEEE Int. Conf. on Neural Networks, pp. 708-712 (1995).
[5] Jun, Y. P., Yoon, H. and Cho, J. W., “L Learning: a Fast Self-Organizing Feature Map Learning Algorithm Based on Incremental Ordering,” IEICE Trans. on Information & Systems, Vol. E76, pp. 698-706 (1993).
[6] Kiang, M. Y., Kulkarni, U. R., Goul, M., Philippakis, A., Chi, R. T. and Turban, E., “Improving the Effectiveness of Self-Organizing Map Networks Using a Circular Kohonen Layer,” Proc. of the 30th Hawaii Int. Conf. on System Sciences, pp. 521-529 (1997).
[7] Koh, J., Suk, M. and Bhandarkar, S. M., “A Multilayer Self-Organizing Feature Map for Range Image Segmentation,” Neural Networks, Vol. 8, pp. 67-86 (1995).
[8] Kohonen, T., Self-Organization and Associative Memory, 3rd ed., Springer-Verlag, Berlin, Germany (1989).
[9] Kohonen, T., Self-Organizing Maps, Springer-Verlag, Berlin, Germany (1995).
[10] Kohonen, T., Self-Organizing Maps, Springer-Verlag, New York, U.S.A. (1995).
[11] Kohonen, T., “The Self-Organizing Feature Map,” Pro. of the IEEE, Vol. 78, pp. 1464-1480 (1990).
[12] Kohonen, T., Oja, E., Simula, O., Visa, A. and Kangas, J.,” Engineering Application of the Self-Organizing Map,” Pro. of the IEEE, Vol. 84, pp. 1358-1383 (1996).
[13] Lo, Z. P. and Bavarian, B., “On the Rate of Convergence in Topology Preserving Neural Networks,” Biological Cybernetics, Vol. 65, pp. 55-63 (1991).
[14] Martinetz, T. M. and Schulten, K. J., “Topology Representing Networks,” Neural Networks, Vol. 7, pp. 507-522 (1994).
[15] McInerney, M. and Dhawan, A., “Training the Self-Organizing Feature Map Using Hybrids of Genetic and Kohonen Methods,” IEEE Int. Conf. on Neural Networks, pp. 641-644 (1994).
[16] Ritter, H. J. and Kohonen, T., “Self-Organizing Semantic Maps,” Biological Cybernetics, Vol. 61, pp.241-254 (1989).
[17] Samad, T. and Harp, S. A., “Self-Organization with Partial Data,” Network: Computation in Neural Systems, Vol. 3, pp. 205-212 (1992).
[18] Su, M. C. and Chang, H. T. “Genetic-Algorithm-Based Approach to Self-Organizing Feature Map and its Application in Cluster Analysis,” IEEE Int. Joint Conf. on Neural Networks, pp. 2116-2121 (1995).
[19] Su, M. C. and Chang, H. T., “Fast Self-Organizing Feature Map,” IEEE Trans. on Neural Networks, Vol. 13, pp. 721-733 (2000).
[20] Tsao, E. C., Bezdek, J. C. and Pal, N. R., “Fuzzy Kohonen Clustering Network,” Pattern Recognition, Vol. 27, pp. 757-764 (1994).
[21] Van Hulle, M. M. and Leuven, K. U., “Globally-Ordered Topology-Preserving Maps Achieved with a Learning Rule Performing Local Weight Updates Only,” IEEE Workshop of Neural Networks for Signal Processing, pp. 95-104 (1995).