Abstract
The article presents a review of one-class support vector machine (1-SVM) used when there is not enough data for abnormal technological object’s behavior detection. Investigated are three procedures of the SVM’s kernel parameter evaluation. Two of them are known in literature as the cross validation method and the maximum dispersion method, and the third one is an author-suggested modification of the maximum dispersion method, minimizing the kernel matrix’s entropy. It is shown that for classification without counting training data set ejections the suggested procedure provides the classification’s quality equal to the first one, and with less value of the kernel parameter.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
References
Vapnik, V.N. and Chervonenkis, A.Ya., Teoriya raspoznavaniya obrazov (statisticheskie problemy obucheniya) (Pattern Recognition Theory: Statistical Problems of Learning), Moscow: Nauka, 1974.
Chandola, V., Banerjee, A., and Kumar, V., Anomaly Detection: A Survey, ACM Comput. Surv., 2009, vol. 41, no. 3.
Markou, M. and Singh, S., Novelty Detection: A Review. Part 1: Statistical Approaches, Signal Proces., 2003, vol. 83, pp. 2481–2497.
Markou, M. and Singh, S., Novelty Detection: A Review. Part 2: Neural Network Based Approaches, Signal Proces., 2003, vol. 83, pp. 2499–2521.
Chang, C. and Lin, C., LIBSVM: A Library for Support Vector Machines, URL: https://www.csie. ntu.edu.tw/~cjlin/libsvm/ (accessed date 26.10.2015).
Iverson, D., Martin, R., Schwabacher, M., et al., General Purpose Data-Driven System Monitoring for Space Operations, AIAA Infotech@Aerospace Conf., Washington: AIAA, 2009, AIAA paper 2009–1909.
Schwabacher, M., Oza, N.C., and Matthews, B.L., Unsupervised Anomaly Detection for Liquid-Fueled Rocket Propulsion Health Monitoring, J. Aerospace Comput., Inform., Commun., 2009, vol. 6, no. 7, pp. 464–482.
Scholkopf, B. and Smola, A.J., Learning with Kernels: Support Vector Machines, Regularization, Optimization, and Beyond, Cambridge: MIT Press, 2002.
Scholkopf, B., Smola, A., Williamson, R.C., and Bartlett, P.L., New Support Vector Algorithms, URL: http://www.stat.purdue.edu/~yuzhu/stat598m3/Papers/NewSVM.pdf (accessed date 02.09.2015).
Cristianini, N., Shawe-Taylor, J., Elissee, A., and Kandola, J., On Kernel-Target Alignment, in Advances in Neural Information Processing Systems 14, Dietterich, T.G., Becker, S., and Ghahramani, Z., Eds, Cambridge: MIT Press, 2002.
Gert, R.G., et al., Learning the Kernel Matrix with Semidefinite Programming, J. Machine Learning Res., 2004. vol. 5, pp. 27–72.
Lin, H.-T. and Lin, C.-J., A Study on Sigmoid Kernels for SVM and the Training of Non-PSD Kernels by SMO-type Methods, Technical Report, Department of Computer Sci., National Taiwan Univ., 2003.
Shawe, T.J. and Nello, C.N., Kernel Methods for Pattern Analysis, Cambridge: Cambridge Univ. Press, 2004.
Evangelista, P.F., Embrechts, M.J., and Szymanski, B.K., Some Properties of the Gaussian Kernel for One Class Learning, Proc. Int. Conf. on Artifical Neural; Networks ICANN07, Portugal, September 2007, Berlin: Springer, 2007, vol. 466, part 1, pp. 269–278.
Runarsson, R.T., Unnthorsson, R., and Johnson, T.M., Model Selection in One Class Nu-SVMS Using RBF Kernels, 16th Conf. on Condition Monitoring and Diagnostic Engineering Management, 2003.
Smylie, D., Clarke, G., and Ulrych, T., Analysis of Irregularities in the Erath’s Rotation, in Methods in Computational Physics, Alder, R., et al., Eds., New York: Academic, 1973, vol. 13, pp. 391–430.
Chessboard, URL: http://pages.cs.wisc.edu/~olvi/data/check1.txt (accessed date 26.10.2015).
URL: http://www.csie.ntu.edu.tw/~cjlin/libsvmtools/datasets/binary.html#fourclass (accessed date 23.10.2015).
Kubat, M., Holte, R., and Matwin, S., Learning When Negative Example Abound, Proc. 9th Eur. Conf. on Machine Learning, ECML 97, Prague, 1997.
Author information
Authors and Affiliations
Corresponding author
Additional information
Original Russian Text © A.N. Budynkov, S.I. Masolkin, 2015, published in Problemy Upravleniya, 2015, No. 6, pp. 70–75.
Rights and permissions
About this article
Cite this article
Budynkov, A.N., Masolkin, S.I. The problem of choosing the kernel for one-class support vector machines. Autom Remote Control 78, 138–145 (2017). https://doi.org/10.1134/S0005117917010118
Received:
Published:
Issue Date:
DOI: https://doi.org/10.1134/S0005117917010118