An efficient algorithm to improve the accuracy and reduce the computations of LS-SVM

Document Type : Research Article

Authors

• M. Baymani ¹
• A. Mansoori ²

¹ Department of Computer and Mathematics, Quchan University of Advanced Technology, Quchan, Iran.

² Department of Applied Mathematics, Ferdowsi University of Mashhad, Mashhad, Iran.

10.22067/ijnao.v10i1.75061

Abstract

We present a novel algorithm, which is called Cutting Algorithm (CA), for improving the accuracy and reducing the computations of the Least Squares Support Vector Machines (LS-SVMs). The method is based on dividing the original problem to some subproblems. Since a master problem is converted to some small problems, so this algorithm has fewer computations. Although, in some cases that the typical LS-SVM cannot classify the dataset linearly, applying the CA the datasets can be classified. In fact, the CA improves the accuracy and reduces the computations. The reported and comparative results on some known datasets and synthetics data demonstrate the efficiency and the performance of CA.

Keywords

• Least squares support vector machine
• Cutting algorithm
• Classification

20.1001.1.24236977.2020.10.1.3.3

---

References

[1] Baymani, M., Salehi-M N. and Mansoori, A. Applying norm concepts for solving interval support vector machine, Neurocomputing 311 (2018), 41–50.

[2] Blake C.L. and Merz, C.J. UCI repository for machine learning databases, Dept. Inf. Comput. Sci., Univ. California, Irvine [online].

[3] Calisir, D. and Dogantekin, E. A new intelligent hepatitis diagnosis system: PCA-LSSVM, Expert Systems with Applications 38 (8) (2011), 10705–10708.

[4] Gao, Y., Shan, X.,Hu, Z., Wang, D., Li, Y. and Tian, X. Extended compressed tracking via random projection based on MSERs and online LS-SVM learning, Pattern Recognit. 59 (2016), 245–254.

[5] Hao, P.-Y. New support vector algorithms with parametric insensitive/margin model, Neural Networks, 23(1) (2010), 60–73.

[6] Kumar M.A. and Gopal, M. Least squares twin support vector machines for pattern classification, Expert Syst. Appl. 36 (2009), 7535–7543.

[7] Lin, C.J. A formal analysis of stopping criteria of decomposition methods for support vector machines, IEEE Transaction on Neural Networks, 13(2002), 1045–1052.

[8] Long, B., Xian, W., Li, M.and Wang, H. Improved diagnostics for the incipient faults in analog circuits using LSSVM based on PSO algorithm with mahalanobis distance, Neurocomputing 133 (10) (2014), 237–248.

[9] Mangasarian O.L. and Wild, E.W. Multisurface proximal support vector classification via generalized eigenvalues, IEEE Transactions on Pattern Analysis and Machine Intelligence, 28 (1), (2006), 69–74.

[10] Mehrkanoon, S. and Suykens, J.A. Learning solutions to partial differen tial equations using LS-SVM, Neurocomputing 159 (2) (2015) 105–116.

[11] Newman, C.B.D. and Merz, C., 1998. UCI repository of machine learning databases. URL http://www.ics.uci.edu/∼mlearn/MLRepository.html

[12] Schölkopf, B., Smola, A., Williamson, R.C. and Bartlett, P.L. New support vector algorithms, Neural Comput. 12 (2000), 1207–1245.

[13] Suykens, J.A.K. and Vandewalle, J. Least squares support vector machine classifiers, Neural Processing Letters 9(3) (1999) 293–300.

[14] Vapnik, V.N. The nature of statistical learning theory, Springer, New York, 1996.

[15] Vapnik, V.N. Statistical learning theory, John Wiley and Sons, New York,1998.

[16] Yang, L.,Yang, S., Li, S., Zhang, R., Liu, F. and Jiao, L. Coupled com pressed sensing inspired sparse spatial-spectral LSSVM for hyperspectral image classification, Knowl-Based Syst. 79 (2015) 80–89.