Center selection of RBF neural network based on modified k-means algorithm with point symmetry distance measure

• Autorzy: Sing J. K. , Basu D. K. , Nasipuri M. , Kundu M.
• Języki publikacji: EN

Abstrakty

EN

The hidden layer neurons of a radial basis function (RBF) neural network map input patterns from a nonlinearly separable space to a linearly separable space. To locate the centers of those hidden layer neurons, normally k-means clustering algorithm is used. Normal k-means clustering algorithm cannot detect hyper spherical-shaped clusters along the principal axes. In present study, we propose a modified version of the k-means clustering algorithm to select RBF centers, which can eliminate this drawback. In the proposed algorithm, we modify the k-means algorithm in two stages. In trie first stage, the procedure to select the initial cluster centers has been modified to capture more knowledge about the distribution of input patterns. In the second stage, the initial centers, selected in the first stage are updated using point symmetry distance measure instead of using conventional Euclidean distance. The RBF neural network with the proposed algorithm has been tested with three different machine-learning data sets. It has also been applied for the segmentation of medical images. The experimental results show that the RBF neural network using the proposed modified k-means algorithm performs better than that using normal k-means algorithm.

Słowa kluczowe

EN

neural network; radial basis function; k-means algorithm

PL

sieć neuronowa

• Wydawca: Wydawnictwo Politechniki Poznańskiej
• Czasopismo: Foundations of Computing and Decision Sciences
• Rocznik: 2004
• Tom: Vol. 29, No. 3
• Strony: 247--266
• Opis fizyczny: Bibliogr. 20 poz.

Twórcy

autor

Sing J. K.

• Dept. of Computer Science & Engg, Jadavpur University, Kolkata 700032, India

autor

Basu D. K.

• Dept. of Computer Science & Engg, Jadavpur University, Kolkata 700032, India

autor

Nasipuri M.

• Dept. of Computer Science & Engg, Jadavpur University, Kolkata 700032, India

autor

Kundu M.

• Dept. of Computer Science & Engg, Jadavpur University, Kolkata 700032, India

Bibliografia

• [1] Blake C., Repository of machine learning databases, Univ. of California, Irvine. [online] http://www.ics.uci.edu/~mleam/MLRepositorv.html.
• [2] Looney C. G., Pattern Recognition Using Neural Networks: theory and algorithms for engineers and scientists, Oxford University Press, NY, 1997.
• [3] Cocosco C.A., Kollokian V., Kwan R.K.-S., Evans A.C.: BrainWeb: Online Interface to a 3D MRI Simulated Brain Database Neurolmage, 5, 4. Available at: http: //www. bic.mni.mcgill.ca/brainweb/, 1997.
• [4] Bors G. and Gabbouj M., "Minimal topology for a radial basis functions neural networks for pattern classification". Digital Processing, 4, 1994, 173-188.
• [5] Zimmermann H. J., Fuzzy Set Theory and its Applications, Kluwer Academic Publications,2nd Ed., 1992.
• [6] Gomm J. B. and Yu D. L., "Selecting radial basis function network centers with recursive orthogonal least squares training", IEEE Trans. Neural Networks, 11, 2000, 306-314.
• [7] Bezdek J., Pattern Recognition with Fuzzy Objective Function Algorithms. New York, Plenum Press, 1991.
• [8] Sing J. K., Basu D. K., Nasipuri M. and Kundu M., "Pattern classification and recognition using generalized radial basis function network". Proceedings of the International Conference on Information Technology, Gopalpur-on-sea, Orissa, India, December 20-22, 2001, 255-259.
• [9] Mao K. Z., "RBF neural network center selection based on fisher ratio class separability measure", IEEE Trans. Neural Networks, 13, 5, 2002, 1211-1217.
• [10] Fu L. M., Neural Networks in Computer Intelligence, McGraw-Hill Inc. 1994.
• [11] Hassoun M. H., Fundamental of Artificial Neural Networks, Cambridge, MA: MIT Press, 1995.
• [12] Joo Er M., Wu S., Lu J. and Toh H. L., "Face recognition with radial basis function (RBF) neural networks", IEEE Trans. Neural Networks, 13, 3, 2002, 697-710.
• [13] Su M-C., Chou C-H, "A modified version of the k-means algorithm with a distance based on cluster symmetry", IEEE Trans. Pattern Analysis and Machine Intelligence, 23, 6, 2001,674-680.
• [14] Bose N. K., Liang P., Neural Network Fundamentals with Graphs, Algorithms, and Applications, McGraw-Hill International Edition, 1996.
• [15]Schokopf R., Sung K. K., Burges C. J. C., Girosi F., Niyogi P., Poggio T., and Vapnik V. N., "Comparing support vector machines with Gaussian kernels to radial basis function classifiers", IEEE Trans. Signal Processing, 45, 1997, 2758-2765.
• [16] Chen S., Cowan C. F., and Grant P. M., "Orthogonal least squares learning algorithms for radial basis function networks", IEEE Trans. Neural networks, 2, 1991, 302-309.
• [17] Haykin S., Neural Networks a Comprehensive Foundation, Prentice-Hall Inc. 2nd Ed. 1999.
• [18] Theodoridis S., Koutroumbas K., Pattern Recognition, Academic Press, 1998.
• [19] Pedrycz W., "Conditional fuzzy clustering in the design of radial basis function neural networks", IEEE, Trans. Neural Networks, 9, 4, 1998, 601-612.
• [20] Hwang Y-S. and Bang S-Y., "Recognition of unconstrained handwritten numerals by a radial basis function network classifier", Pattern Recognition Letters, 18, 1997, 657-664.