Fusion of Self-Organizing Maps with Different Sizes
An ensemble consists of several neural networks whose outputs are fused to produce a single output, which usually will be better than the individual results of each network. This work presents a methodology to aggregate the results of several Kohonen Self-Organizing Maps in an ensemble. Computational simulations demonstrate an increase in the accuracy classification and the proposed method effectiveness was evidenced by the Wilcoxon Signed Rank Test.
KeywordsEnsemble Self-Organizing Maps Classification
Authors would like to thank the support of CAPES Foundation, Ministry of Education of Brazil, Brasilia - DF, Zip Code 70.040-020.
- 2.Perrone, M.P., Cooper, L.N.: When networks disagree: ensemble methods for hybrid neural networks. In: Mammone, R.J. (ed.) Neural Networks for Speech and Image Processing, pp. 126–142. Chapman and Hall, New York (1993)Google Scholar
- 4.Gonalves, M.L., De Andrade Netto, M.L., Costa, J.A.F., Zullo, J.: Data clustering using self-organizing maps segmented by mathematic morphology and simplified cluster validity indexes: an application in remotely sensed images. In: IJCNN 2006, International Joint Conference on Neural Networks, pp. 4421–4428 (2006)Google Scholar
- 5.Georgakis, A., Li, H., Gordan, M.: An ensemble of SOM networks for document organization and retrieval. In: International Conference on Adaptive Knowledge Representation and Reasoning (2005)Google Scholar
- 11.DeLooze, L.L.: Attack characterization and intrusion detection using an ensemble of self-organizing maps. In: 2006 IEEE Information Assurance Workshop, pp. 108–115 (2006)Google Scholar
- 15.Bache, K., Lichman, M.: Machine Learning Repository. University of California, School of Information and Computer Sciences, Irvine (2013) Google Scholar
- 16.Ultsch, A.: Clustering with SOM: U*C. In: Proceedings Workshop on Self-Organizing Maps (WSOM 2005), pp. 75–82 (2005)Google Scholar