Abstract
This paper presents parameter tuning of multi-class support vector machines (SVMs) using radius–margin bound through data envelopment analysis (DEA). Relative efficiencies of the binary SVMs in one-vs-one multi-class SVM are obtained using radius as the input and margin as the output by employing Charnes, Cooper and Rhodes model (CCR model). The objective is to maximize the sum of the relative efficiencies of individual binary SVMs in one-vs-one multi-class SVM. The proposed formulation is empirically found to be outperforming Keerthi’s radius–margin based parameter tuning. In the best case, the radius obtained through the proposed method is 1,110 times lower than Keerthi’s method. Margin is 4.4 times higher, the number of support vectors are 5.8 times lower and test error rates are competitive. Proposed method is also compared with (i) Widely employed grid search method and (ii) Parameter tuning method proposed in (IEEE Trans Mach Intell 32:1888–1898, 2010. The merit of the proposed method over these tuning methods is shown through experimentation along the four identified dimensions.
Similar content being viewed by others
Notes
All the input and outputs are assumed to be positive.
We thank the authors of the paper [8] for sharing their code through their webpage. The code can be downloaded at:www.dsp.rice.edu/software.
References
Allwein EL, Schapire RE, Singer Y (2000) Reducing multiclass to binary: a unifying approach for margin classifiers. J Mach Learn Res 1:113–141
Asuncion A, Newman D (2007) UCI machine learning repository. http://www.ics.uci.edu/mlearn/MLRepository.html
Burges CJC (1998) A tutorial on support vector machines for pattern recognition. Data Min Knowl Discov 2(2):121–167
Chapelle O, Vapnik V, Bousquet O, Mukherjee S (2002) Choosing multiple parameters for support vector machines. Mach Learn 46:131–159
Charnes A, Cooper W, Rhodes E (1978) Measuring the efficiency of decision making units. Eur J Oper Res 2(6):429–444. doi:10.1016/0377-2217(78)90138-8
Cooper WW, Seiford LM, Tone K (2000) Data envelopment analysis: a comprehensive text with models, applications, references and DEA-solver softwares. Kluwer Academic, New York
Cortes C, Vapnik VN (1995) Support vector networks. Mach Learn 20:273–297
Davenport MA, Baraniuk RG, Scott CD (2010) Tuning support vector machines for minimax and Neyman–Pearson classification. IEEE Trans Pattern Anal Mach Intell 32(10):1888–1898
Dietterich TG, Bakiri G (1995) Solving multiclass learning problems via error-correcting output codes. J Artif Intell Res 2:263–286
Friedman J (1996) Another approach to polychotomous classification. Online technical report. Stanford University, Stanford
Friedrichs F, Igel C (2004) Evolutionary tuning of multiple SVM parameters. Neurocomputing 64:3326–3334
Fürnkranz J (2002) Round robin classification. J Mach Learn Res 2:721–747
Goldberg DE (1989) Genetic algorithms in search optimization and machine learning. Addison-Wesley, Bosto
Hsu CW, Lin CJ (2002) A comparison of methods for multi-class support vector machines. IEEE Trans Neural Netw 13:415–425
Hsu CW, Chang CC, Lin CJ (2010) A practical guide to support vector classification. Bioinformatics 1:1–16
Joachims T (2000) The maximum margin approach to learning text classifiers: methods, theory, and algorithms. PhD thesis, Universität Dortmund, Informatik, LS VIII.
Keerthi S (2002) Efficient tuning of SVM hyperparameters using radius/margin bound and iterative algorithms. Neural Netw IEEE Trans 13(5):1225–1229. doi:10.1109/TNN.2002.1031955
Lanckriet GRG, Cristianini N, Bartlett P, Ghaoui LE, Jordan MI (2004) Learning the kernel matrix with semidefinite programming. J Mach Learn Res 5:27–72
Lee Y, Lin Y, Wahba G (2001a) Multicategory support vector machines. Tech. rep., Department of Statistics, University of Wisconsin.
Lee Y, Lin Y, Wahba G (2001b) Multicategory support vector machines. In The 33rd symposium on the Interface.
Lins MPE, Gomes EG, de Mello JCCS, de Mello AJRS (2003) Olympic ranking based on a zero sum gains DEA model. Eur J Oper Res 148:312–322
Lorena AC, de Carvalho AC (2008) Evolutionary tuning of SVM parameter values in multiclass problems. Neurocomputing 71:3326–3334
Murphy PM, Aha DW (1993) UCI repository of machine learning databases (machine-readable data repository). Tech. rep., Department of Information and Computer Science, University of California, Irvine. http://www.ics.uci.edu/~mlearn/MLRepository.html
Saha A, Ravisankar TS (2000) Rating of Indian commercial banks: a DEA approach. Eur J Oper Res 124:187–203
Schölkopf B, Burges C, Vapnik V (1995) Extracting support data for a given task. In First international conference on knowledge discovery and data mining, New York.
Shanno DF, Phu KH (1980) Minimization of unconstrained multivariate functions. ACM Trans Math Softwar 2(1):87–94
Staub RB, da Silva e Souza G, Tabak BM (2010) Evolution of bank efficiency in Brazil: a DEA approach. Eur J Oper Res 202:204–213
Stensrud E, Myrtveit I (2003) Identifying high performance ERP projects. IEEE Trans Softw Eng 29(5):398–416
Valentini G, Dietterich TG (2006) Bias-variance analysis of support vector machines for the development of SVM-based ensemble methods. J Mach Learn Res 5:725–775
Vapnik V, Chapell O (2000) Bounds on error expectation for support vector machines. Neural Comput 12:2013–2036. doi:10.1162/089976600300015042
Vapnik VN (1998) Stat Learn Theory. Wiley, New York
Weston J (1999) Multi-class support vector machines. In Proceedings of ESANN, Brussels D. Facto Press.
Wu J, Liang L, Yang F (2009) Achievement and benchmarking of countries at the summer olympics using cross efficiency evaluation method. Eur J Oper Res 197:722–730
Zhao X, Wang S, Lai KK (2011) Mutual funds performance evaluation based on endogenous benchmarks. Exp Syst Appl 38:3663–3670
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Vijaya Saradhi, V., Girish, K.R. Effective Parameter Tuning of SVMs Using Radius/Margin Bound Through Data Envelopment Analysis. Neural Process Lett 41, 125–138 (2015). https://doi.org/10.1007/s11063-014-9338-9
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11063-014-9338-9