Supervised SOM Based Architecture versus Multilayer Perceptron and RBF Networks

David Gil
Computing Technology and Data Processing, University of Alicante, Spain

Magnus Johnsson
Lund University Cognitive Science, Lund, Sweden

Ladda ner artikel

Ingår i: The Swedish AI Society Workshop May 20-21; 2010; Uppsala University

Linköping Electronic Conference Proceedings 48:5, s. 15-24

Visa mer +

Publicerad: 2010-05-19


ISSN: 1650-3686 (tryckt), 1650-3740 (online)


We address a contrastive study between the well known Multi-Layer Perceptron (MLP) and Radial Basis Function (RBF) neural networks and a SOM based supervised architecture in a number of data classification tasks. Well known databases like Breast Cancer; Parkinson and Iris were used to evaluate the three architectures by constructing confusion matrices. The results are encouraging and indicate that the SOM based supervised architecture generally achieves results as good as the MLP and slightly higher on some measures than the RBF network.


Inga nyckelord är tillgängliga


[1] A. Asuncion and D.J. Newman. UCI machine learning repository; 2007.

[2] C. Balkenius; J. Morén; B. Johansson; and M. Johnsson. Ikaros: Building cognitive mod els for robots. Advanced Engineering Infor- matics; 24(1):40-48; 2010.

[3] C. Bishop. Improving the generalization prop- erties of radial basis function neural networks. Neural Computation; 3(4):579-588; 1991.

[4] CM Bishop. Neural networks for pattern recognition: Oxford University Press. New York; New York; USA; 1995.

[5] C.M. Bishop. Neural networks for pattern recognition. Oxford Univ Pr; 2005.

[6] Fi-John Chang; Li-Chiu Chang; Huey-Shan Kao; and Gwo-Ru Wu. Assessing the e®ort of meteorological variables for evaporation estimation by self-organizing map neural net- work. Journal of Hydrology; In Press; Corrected Proof:; 2010.

[7] B. Choi and J.H. Lee. Comparison of generalization ability on solving differential equations using backpropagation and reformulated radial basis function networks. Neurocomput- ing; 2009.

[8] A.E. Diaz and C.A. Hernandez. Tool for the design and implementation of control systems with neural networks. In IEEE International Conference on Intelligent Computing and In- telligent Systems; 2009. ICIS 2009; volume 1; 2009.

[9] D. Gil; M. Johnsson; J.M. Garcia Chamizo; A.S. Paya; and D.R. Fernandez. Application of artificial neural networks in the diagnosis of urological dysfunctions. Expert Systems with Applications; 36(3P2):5754-5760; 2009.

[10] David Gil and Magnus Johnsson. Using sup- port vector machines in diagnoses of urologi- cal dysfunctions. Expert Systems with Applications; 37(6):4713-4718; 2010.

[11] S. Haykin. Neural Networks: A Comprehensive Foundation; Englewoods Cliffs; 1998.

[12] S. Haykin. Neural networks: a comprehensive foundation. Prentice Hall; 2008.

[13] C.N. Kim; K.H. Yang; and J. Kim. Hu- man decision-making behavior and modeling effects. Decision Support Systems; 45(3):517-527; 2008.

[14] R. Kohavi and F. Provost. Glossary of terms. Machine Learning; 30(2/3):271-274; 1998.

[15] T. Kohonen. Self-Organization and Associative Memory. Springer Verlag; 1988.

[16] M.A. Little; P.E. McSharry; E.J. Hunter; J. Spielman; and L.O. Ramig. Suitability of dysphonia measurements for telemonitoring of Parkinson? s disease. IEEE transactions on bio-medical engineering; 2008.

[17] M.A. Little; P.E. McSharry; S.J. Roberts; D.A.E. Costello; and I.M. Moroz. Exploiting Nonlinear recurrence and Fractal scaling properties for voice disorder detection. BioMedical Engineering OnLine; 6(1):23; 2007.

[18] J.V. Marcos; R. Hornero; D. Alvarez; F. del Campo; M. Lopez; and C. Zamarron. Radial basis function classifiers to help in the diagno- sis of the obstructive sleep apnoea syndrome from nocturnal oximetry. Medical and Bio- logical Engineering and Computing; 46(4):323-332; 2008.

[19] I. MathWorks. MATLAB & SIMULINK In-strument Control Toolbox 2. The MathWorks Inc.; 2007.

[20] S. Mehrabi; M. Maghsoudloo; H. Arabalibeik; R. Noormand; and Y. Nozari. Application of multilayer perceptron and radial basis function neural networks in differentiating between chronic obstructive pulmonary and congestive heart failure diseases. Expert Systems With Applications; 36(3P2):6956-6959; 2009.

[21] M.R. Narasinga Rao; GR Sridhar; K. Madhu; and A.A. Rao. A clinical decision support system using multi-layer perceptron neural net- work to predict quality of life in diabetes. Di- abetes & Metabolic Syndrome: Clinical Research & Reviews; 2009.

[22] M. Pal and University of Nottingham (GB). Factors Influencing the Accuracy of Remote Sensing Classification: A Comparative Study. University of Nottingham; 2002.

[23] P. Piela. Introduction to Self-Organizing Maps Modelling for Imputation-Techniques and Technology. Research in O±cial Statis- tics; 2:5-19; 2002.

[24] B.D. Ripley. Pattern recognition and neural networks. Cambridge university press; 1996.

[25] D. E. Rumelhart; G. E. Hinton; and R. J. Williams. Learning representations by back- propagating errors. Nature; 323:533-536; 1986.

[26] F. Temurtas. A comparative study on thyroid disease diagnosis using neural networks. Ex- pert Systems with Applications; 36(1):944-949; 2009.

[27] H. Werner and M. Obach. New neural network types estimating the accuracy of response for ecological modelling. Ecological Modelling; 146(1-3):289-298; 2001.

[28] H. Yan; Y. Jiang; J. Zheng; C. Peng; and Q. Li. A multilayer perceptron-based medi- cal decision support system for heart disease diagnosis. Expert Systems with Applications; 30(2):272-281; 2006.

Citeringar i Crossref