Abstract
The essential characteristic of constructive neural network (CoNN) algorithms is the incremental construction of the neural network architecture along with the training process. The BabCoNN (Barycentric-based constructive neural network) algorithm is a new neural network constructive algorithm suitable for two-class problems that relies on the BCP (Barycentric Correction Procedure) for training its individual TLU (Threshold Logic Unit). Motivated by the good results obtained with the two-class BabCoNN, this paper proposes its extension to multiclass domains as a new CoNN algorithm named MBabCoNN. Besides describing the main concepts involved in the MBabCoNN proposal, the paper also presents a comparative analysis of its performance versus the multiclass versions of five well known constructive algorithms, in four knowledge domains as an empirical evidence of the MBabCoNN suitability and efficiency for multiclass classification tasks.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsPreview
Unable to display preview. Download preview PDF.
References
Duda, R.O., Hart, P.E., Stork, D.G.: Pattern Classification. John Wiley, Chichester (2001)
Mitchell, T.M.: Machine Learning. McGraw-Hill, New York (1997)
Nilsson, N.J.: Learning Machines. McGrall-Hill Systems Science Series (1965)
Elizondo, D.: The linear separability problem: Some testing methods. IEEE Transactions on Neural Network 17(2), 330–344 (2006)
Bertini Jr., J.R., Nicoletti, M.C.: A constructive neural network algorithm based on the geometric concept of barycenter of convex hull. In: ICAISC (June 2008) (accepted for presentation)
Gallant, S.I.: Neural Network Learning & Expert Systems. MIT Press, Cambridge (1994)
Mézard, M., Nadal, J.: Learning feedforward networks: The tiling algorithm. J. Phys. A: Math Gen. 22, 2191–2203 (1989)
Frean, M.: The upstart algorithm: a method for constructing and training feedforward neural networks. Neural Computation 2, 198–209 (1990)
Amaldi, E., Guenin, B.: Two constructive methods for designing compact feedforward networks of threshold units. Int. Journal of Neural System 8(5), 629–645 (1997)
Nicoletti, M.C., Bertini Jr., J.R.: An empirical evaluation of constructive neural network algorithms in classification tasks. Int. Journal of Innovative Computing and Applications (IJICA) 1, 2–13 (2007)
Poulard, H.: Baricentric correction procedure: A fast method for learning threshold unit. In: WCNN, vol. 1, pp. 710–713 (1995)
Poulard, H., Labreche, S.: A new threshold unit learning algorithm. Technical Report 95504, LAAS (1995)
Bertini Jr., J.R., Nicoletti, M.C., Hruschka Jr., E.R.: A comparative evaluation of constructive neural networks methods using prm and bcp as tlu training algorithms. In: Proc. of the IEEE Int. Conf. Systems, Man, and Cybernetics, Taiwan, pp. 3497–3502 (2006)
Poulard, H., Estèves, D.: A convergence theorem for barycentric correction procedure. Technical Report 95180, LAAS-CNRS (1995)
Parekh, R.G., Yang, J., Honovar, V.: Constructive neural network learning algorithm for multi-category classification. Technical Report TR ISU-CS-TR95-15a, Iowa State University (1995)
Parekh, R.G., Yang, J., Honovar, V.: Mupstart - a constructive neural network learning algorithm for multi-category pattern classification. In: ICNN, vol. 3, pp. 1920–1924(1997)
Yang, J., Parekh, R., Honovar, V.: Mtiling - a constructive network learning algorithm for multi-category pattern classification. In: World Congress on Neural Networks, 182–187 (1996)
Burgess, N.: A constructive algorithm that converges for real-valued input patterns. International Journal of Neural Systems 5(1), 59–66 (1994)
Fahlman, S., Lebiere, C.T.: The Cascade Correlation Architecture. In: Advances in Neural Information Processing Systems 2, pp. 524–532. Morgan Kaufman, San Francisco (1990)
Asuncion, A., Newman, D.J.: Uci machine learning repository. University of California, School of Information and Computer Science, Irvine, CA (2007)
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 2008 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Bertini, J.R., do Carmo Nicoletti, M. (2008). MBabCoNN – A Multiclass Version of a Constructive Neural Network Algorithm Based on Linear Separability and Convex Hull. In: Kůrková, V., Neruda, R., Koutník, J. (eds) Artificial Neural Networks - ICANN 2008. ICANN 2008. Lecture Notes in Computer Science, vol 5164. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-540-87559-8_75
Download citation
DOI: https://doi.org/10.1007/978-3-540-87559-8_75
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-87558-1
Online ISBN: 978-3-540-87559-8
eBook Packages: Computer ScienceComputer Science (R0)