论文信息 - Advantages and drawbacks of the Batch Kohonen algorithm

Advantages and drawbacks of the Batch Kohonen algorithm

The Kohonen algorithm (SOM) was originally defined as a stochastic algorithm which works in an on-line way and which was designed to model some plastic features of the human brain. In fact it is nowadays extensively used for data mining, data visualization, and exploratory data analysis. Some users are tempted to use the batch version of the Kohonen algorithm (KBATCH) since it is a deterministic algorithm which can go faster in some cases. After [7], which tried to elucidate the mathematical nature of the Batch variant, in this paper, we give some elements of comparison for both algorithms, using theoretical arguments, simulated data and real data.

Marie Cottrell | Patrick Letrémy | Jean-Claude Fort

[1] J.-C. Fort,et al. Stochastic on-line algorithm versus batch algorithm for quantization and self organizing maps , 2001, Neural Networks for Signal Processing XI: Proceedings of the 2001 IEEE Signal Processing Society Workshop (IEEE Cat. No.01TH8584).

[2] Gilles Pagès,et al. Theoretical aspects of the SOM algorithm , 1998, Neurocomputing.

[3] Michel Verleysen,et al. Using the Kohonen algorithm for quick initialization of Simple Competitive Learning algorithm , 1999, ESANN.

[4] Settimo Termini,et al. T. Kohonen,self-organizing maps , 1995 .

[5] Patrick Rousset,et al. The Kohonen Algorithm: A Powerful Tool for Analyzing and Representing Multidimensional Quantitative and Qualitative Data , 1997, IWANN.

[6] T. Kohonen,et al. Visual Explorations in Finance with Self-Organizing Maps , 1998 .

[7] Helge J. Ritter,et al. Neural computation and self-organizing maps - an introduction , 1992, Computation and neural systems series.

[8] E. Forgy,et al. Cluster analysis of multivariate data : efficiency versus interpretability of classifications , 1965 .

[9] Teuvo Kohonen,et al. Comparison of SOM Point Densities Based on Different Criteria , 1999, Neural Computation.

[10] Teuvo Kohonen,et al. Self-Organization and Associative Memory , 1988 .