Abstract
Model selection plays a key role in the performance of support vector machines (SVMs). At present, nearly all researches are based on binary classification and focus on how to estimate the generalization performance of SVMs effectively and efficiently. For problems with more than two classes, where a classifier is typically constructed by combining several binary SVMs [8], most researchers simply select all binary SVM models simultaneously in one hyper-parameter space. Though this all-in-one method works well, there is another choice – the one-in-one method where each binary SVM model is selected independently and separately. In this paper, we compare the two methods for multi-class SVMs with the one-against-one strategy [8]. Their properties are discussed and their performance is analyzed based on experimental results.
This work is supported by the National Natural Science Foundation of China under grant No. 60072029 and No.60271033.
This is a preview of subscription content, log in via an institution to check access.
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
Burges, C.J.: A Tutorial on Support Vector Machines for Pattern Recognition. Data Mining and Knowledge Discovery 2, 121–267 (1998)
Chung, K.-M., Kao, W.-C., Sun, T., Wang, L.-L., Lin, C.-J.: Radius Margin Bounds for Support Vector Machines with the RBF Kernel. Neural Computation 11, 2643–2681 (2003)
Chang, C.-C., Lin, C.-J.: LIBSVM: A Library for Support Vector Machines (2002), Online at http://www.csie.ntu.edu.tw/~cjlin/papers/libsvm.pdf
Staelin, C.: Parameter Selection for Support Vector Machines (2003), Online at http://www.hpl.hp.com/techreports/2002/HPL-2002-354R1.pdf
Li, H.-Q., Wang, S.-Y., Qi, F.-H.: Minimal Enclosing Sphere Estimation and Its Application to SVMs Model Selection. In: Yin, F.-L., Wang, J., Guo, C.G. (eds.) ISNN 2004. LNCS, vol. 3173, pp. 487–493. Springer, Heidelberg (2004)
Chapelle, O., Vapnik, V., Bousquet, O., Mukherjee, S.: Choosing Multiple Parameters for Support Vector Machines. Machine Learning 46, 131–159 (2002)
Kressel, U.: Pairwise Classification and Support Vector Machines. In: Schölkopf, B., Burges, C., Smola, A. (eds.) Advances in Kernel Methods: Support Vector Learning, pp. 255–268. MIT Press, Cambridge (1999)
Hsu, C.-W., Lin, C.-J.: A comparison of methods for multi-class support vector machines. IEEE Trans. on Neural Networks 13, 415–425 (2002)
Wu, T.-F., Lin, C.-J., Weng, R.C.: Probability Estimates for Multi-Class Classification by Pairwise Coupling. Journal of Machine Learning Research 5, 975–1005 (2004)
Blake, C.L., Merz, C.J.: UCI Repository of Machine Learning Databases (1998), Online at http://www.ics.uci.edu/~mlearn/MLRepository.html
Michie, D., Spiegelhalter, D.J., Taylor, C.C.: Machine Learning, Neural and Statistical Classification. Ellis Horwood, London (1994), Data available at ftp://ftp.ncc.up.pt/pub/statlog
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2005 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Li, H., Qi, F., Wang, S. (2005). A Comparison of Model Selection Methods for Multi-class Support Vector Machines. In: Gervasi, O., et al. Computational Science and Its Applications – ICCSA 2005. ICCSA 2005. Lecture Notes in Computer Science, vol 3483. Springer, Berlin, Heidelberg. https://doi.org/10.1007/11424925_119
Download citation
DOI: https://doi.org/10.1007/11424925_119
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-25863-6
Online ISBN: 978-3-540-32309-9
eBook Packages: Computer ScienceComputer Science (R0)