Abstract
Finding a kernel mapping function is a key step towards construction of a high-performanced SVM-based classifier. While some recent methods exploited an evolutional approach to construct a suitable multifunction kernel, most of them searched randomly and diversely. In this paper, the concept of a family of identical-structured kernel trees is proposed to enable exploration of structure space using genetic programming whereas to pursue investigation of parameter space on a certain tree using evolutional strategy. To control balance between structure and parameter search towards an optimal kernel, the trade-off strategy is introduced. By experiments on a number of benchmark datasets from UCI and text classification datasets, the proposed method is shown to be able to find a better optimal solution than other search methods.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Chapelle, O., Vapnik, V., Bousquet, O., Mukherjee, S.: Choosing multiple param- eters for support vector machines. Mach. Learn. 46(1-3), 131–159 (2002)
Keerthi, S.: Efficient tuning of svm hyperparameters using radius/margin bound and iterative algorithms. IEEE Transactions on Neural Networks 13, 1225–1229 (2002)
Lanckriet, G.R.G., Cristianini, N., Bartlett, P., Ghaoui, L.E., Jordan, M.I.: Learn- ing the kernel matrix with semidefinite programming. J. Mach. Learn. Res. 5, 27–72 (2004)
Sonnenburg, S., Räatsch, G., Schäafer, C., Schölkopf, B.: Large Scale Multiple Kernel Learning. Journal of Machine Learning Research 7, 1531–1565 (2006)
Phienthrakul, T., Kijsirikul, B.: Evolutionary strategies for multi-scale radial basis function kernels in support vector machines. In: Proceedings of the 2005 conference on Genetic and evolutionary computation, pp. 905–911. ACM Press, New York (2005)
Sullivan, K.M., Luke, S.: Evolving kernels for support vector machine classification. In: Proceedings of the 9th annual conference on Genetic and evolutionary computation, pp. 1702–1707. ACM, New York (2007)
Methasate, I., Theeramunkong, T.: Experiments on kernel tree support vector machines for text categorization. In: Zhou, Z.-H., Li, H., Yang, Q. (eds.) PAKDD 2007. LNCS, vol. 4426, pp. 720–727. Springer, Heidelberg (2007)
Methasate, I., Theeramunkong, T.: Kernel trees for support vector machines. IE- ICE Transactions 90-D(10), 1550–1556 (2007)
Phienthrakul, T., Kijsirikul, B.: GPES: An algorithm for evolving hybrid kernel functions of support vector machines. In: 2007 IEEE Congress on Evolutionary Computation, Singapore, September 25-28, pp. 2636–2643. IEEE Press, Los Alamitos (2007)
Ghani, R.: Cmu world wide knowledge base (webkb) project (2001)
Lang, K.: Newsweeder: Learning to filter netnews. In: Proceedings of the Twelfth International Conference on Machine Learning, pp. 331–339 (1995)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2009 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Methasate, I., Theeramunkong, T. (2009). A Family-Based Evolutional Approach for Kernel Tree Selection in SVMs. In: Theeramunkong, T., Kijsirikul, B., Cercone, N., Ho, TB. (eds) Advances in Knowledge Discovery and Data Mining. PAKDD 2009. Lecture Notes in Computer Science(), vol 5476. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-01307-2_111
Download citation
DOI: https://doi.org/10.1007/978-3-642-01307-2_111
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-01306-5
Online ISBN: 978-3-642-01307-2
eBook Packages: Computer ScienceComputer Science (R0)