Polyhedral conic kernel-like functions for SVMs

In this study, we propose a new approach that can be used as a kernel-like function for support vector machines(SVMs) in order to get nonlinear classification surfaces. We combined polyhedral conic functions (PCFs) with the SVM method. To get nonlinear classification surfaces, kernel functions are used with SVMs. However, the parameter selection of the kernel function affects the classification accuracy. Generally, in order to get successful classifiers which can predict unknown data accurately, best parameters are explored with the grid search method which is computationally expensive. We solved this problem with the proposed method. There is no need to optimize any parameter in the proposed method. We tested the proposed method on three publicly available datasets. Next, the classification accuracies of the proposed method were compared with the linear, radial basis function (RBF), Pearson universal kernel (PUK), and polynomial kernel SVMs. The results are competitive with those of the other methods.

PDF

___

[1] Gasimov R, Ozturk G. Separation via polyhedral conic functions. Optimization Methods & Software 2006; 21(4): 527-540.
[2] Uylas Sati N. A binary classification approach based on support vector machines via polyhedral conic functions. Celal Bayar University Journal of Science 2016; 12(2): 135-149.
[3] Cortes C, Vapnik V. Support-vector networks. Machine Learning 1995; 20(3): 273-297.
[4] Boser BE, Guyon IM, Vapnik VN. A training algorithm for optimal margin classifiers. In: Proceedings of the Fifth Annual Workshop on Computational Learning Theory, COLT ’92. ACM, New York, NY, USA, 1992; 144-152.
[5] Lodhi H, Saunders C, Shawe-Taylor J, Cristianini N, Watkins C. Text classification using string kernels. Journal of Machine Learning Research 2002; 2: 419-444.
[6] Schölkopf B, Tsuda K, Vert JP. Kernel Methods in Computational Biology. Cambridge, Massachusetts, London, England: MIT Press, 2004.
[7] Gönen M, Alpaydın E. Multiple kernel learning algorithms. Journal of Machine Learning Research 2011; 12: 2211-2268.
[8] Boser BE, Guyon IM, Vapnik VN. A training algorithm for optimal margin classifiers. In: Proceedings of the fifth annual workshop on Computational learning theory. ACM, 1992; 144-152.
[9] Hsu CW, Chang CC, Lin CJ. A practical guide to support vector classification. Technical report, Department of Computer Science, National Taiwan University, 2003.
[10] Bagirov AM, Ugon J, Webb D, Ozturk G, Kasimbeyli R. A novel piecewise linear classifier based on polyhedral conic and max–min separabilities. Top 2013; 21(1): 3-24.
[11] Ozturk G, Ciftci MT. Clustering based polyhedral conic functions algorithm in classification. Journal of Industrial and Management Optimization 2015; 11(3): 921-932.
[12] Ozturk G, Bagirov AM, Kasimbeyli R. An incremental piecewise linear classifier based on polyhedral conic separation. Machine Learning 2015; 101(1-3, SI): 397-413.
[13] Cimen E, Ozturk G. Arrhythmia classification via k-Means based polyhedral conic functions algorithm. In: Arabnia, HR and Deligiannidis, L and Yang, M, editor, 2016 International Conference on Computational Science & Computational Intelligence (CSCI). Amer Council Sci & Educ, 2016; 798-802. International Conference on Computational Science and Computational Intelligence (CSIC), Las Vegas, NV, USA, 15–17 December, 2016.
[14] Ozturk G, Ceylan G. Max margin polyhedral conic function classifier. In: Arabnia, HR and Deligiannidis, L and Yang, M, editor, 2016 International Conference on Computational Science & Computational Intelligence (CSCI). Amer Council Sci & Educ, 2016; 1395-1396. International Conference on Computational Science and Computational Intelligence (CSIC), Las Vegas, NV, USA, 15–17 December, 2016.
[15] Cimen E, Ozturk G, Gerek O N. Incremental conic functions algorithm for large scale classification problems. Digital Signal Processing 2018; 77: 187-194.
[16] Cimen E, Ozturk G, Gerek ON. Icf: an algorithm for large scale classification with conic functions. SoftwareX 2018; 8: 59-63.
[17] Cevikalp H, Triggs B. Polyhedral conic classifiers for visual object detection and classification. In: 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). 2017; 4114-4122.
[18] Cimen E. Gesture Recognition with Polyhedral Conic Functions based Classifiers. MSc, Graduate School of Science, Anadolu University, 2013 (in Turkish).
[19] Sati NU. A binary classification algorithm based on polyhedral conic functions. Düzce Üniversitesi Bilim ve Teknoloji Dergisi 2015; 3(1): 152-161.
[20] Sati NU, Ordin B. Application of the polyhedral conic functions method in the text classification and comparative analysis. Scientific Programming 2018; 2018: 1-11.
[21] Sarıbaş H, Çevıkalp H, Kahvecıoğlu S. Car detection in images taken from unmanned aerial vehicles. In: 2018 26th Signal Processing and Communications Applications Conference (SIU). IEEE, 2018; 1-4.
[22] MacQueen J. Some methods for classification and analysis of multivariate observations. In: Proceedings of the Fifth Berkeley Symposium on Mathematical Statistics and Probability, Volume 1: Statistics. University of California Press, Berkeley, CA, USA. pp. 281-297.
[23] Chang CC, Lin CJ. Libsvm: A library for support vector machines. ACM Trans Intell Syst Technol 2011; 2(3): 1-27.
[24] Gurobi Optimization L. Gurobi optimizer reference manual, 2018.
[25] Hall M, Frank E, Holmes G, Pfahringer B, Reutemann P, Witten IH. The weka data mining software: an update. ACM SIGKDD Explorations Newsletter 2009; 11(1): 10-18.