Abstract
In pattern classification, the geometric method often provides a simple and intuitive solution. In the case of linear separability, solving the optimization hyperplane problem can be transformed into solving the nearest point problem of the convex hulls between classes. In the case of nonlinear separability, the notion of scaled convex hull (SCH) is employed to reduce the initially overlapping convex hulls to become separable. Two classic nearest point algorithms, GSK and MDM, have been used as effective solvers for SCHs. However, their problem-solving speed is still a bit underperforming. This paper proposes a new solver called SCH-CDM, in which the CDM (cross distance minimization) algorithm is employed to calculate the nearest point pair of between-class SCHs. Experimental results indicate that the SCH-CDM algorithm can achieve faster convergence than the SCH-GSK algorithm and the SCH-MDM algorithm. In terms of accuracy, it also shows good competitiveness compared to the baseline methods.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
References
Peng, X., Wang, Y.: Geometric algorithms to large margin classifier based on affine hulls. IEEE Trans. Neural Netw. Learn. Syst. 23(2), 236–246 (2011)
Leng, Q., Wang, S., Qin, Y., et al.: An effective method to determine whether a point is within a convex hull and its generalized convex polyhedron classifier. Inf. Sci. 504, 435–448 (2019)
Ding, S., Nie, X., Qiao, H., et al.: A fast algorithm of convex hull vertices selection for online classification. IEEE Trans. Neural Netw. Learn. Syst. 29(4), 792–806 (2017)
Carmichael, I., Marron, J.S.: Geometric insights into support vector machine behavior using the KKT conditions. Electron. J. Stat. 15(2), 6311–6343 (2021)
Chen, X., Xiao, Y.: Geometric projection twin support vector machine for pattern classification. Multimedia Tools Appl. 80(15), 23073–23089 (2020). https://doi.org/10.1007/s11042-020-09103-1
Bennett, K.P., Bredensteiner, E.J.: Duality and geometry in SVM classifiers. In: Seventeenth International Conference on Machine Learning (ICML 2000), pp. 57–64 (2000)
Barbero, A., Takeda, A., López, J.: Geometric intuition and algorithms for Eν–SVM. J. Mach. Learn. Res. 16, 323–369 (2015)
Almasi, O.N., Rouhani, M.: A geometric-based data reduction approach for large low dimensional datasets: Delaunay triangulation in SVM algorithms. Mach. Learn. Appl. 4, 100025 (2021)
Franc, V., Hlaváč, V.: An iterative algorithm learning the maximal margin classifier. Pattern Recogn. 36(9), 1985–1996 (2003)
Keerthi, S.S., Shevade, S.K., Bhattacharyya, C., et al.: A fast iterative nearest point algorithm for support vector machine classifier design. IEEE Trans. Neural Netw. 11(1), 124–136 (2000)
Yujian, L., Bo, L., Xinwu, Y., et al.: Multiconlitron: a general piecewise linear classifier. IEEE Trans. Neural Netw. 22(2), 276–289 (2011)
Mavroforakis, M.E., Theodoridis, S.: A geometric approach to support vector machine (SVM) classification. IEEE Trans. Neural Netw. 17(3), 671–682 (2006)
Mavroforakis, M.E., Sdralis, M., Theodoridis, S.: A novel SVM geometric algorithm based on reduced convex hulls. In: 18th International Conference on Pattern Recognition (ICPR 2006), pp. 564–568 (2006)
Mavroforakis, M.E., Sdralis, M., Theodoridis, S.: A geometric nearest point algorithm for the efficient solution of the SVM classification task. IEEE Trans. Neural Netw. 18(5), 1545–1549 (2007)
López, J., Barbero, Á., Dorronsoro, J.R.: Clipping algorithms for solving the nearest point problem over reduced convex hulls. Pattern Recogn. 44(3), 607–614 (2011)
Liu, Z., Liu, J.G., Pan, C., et al.: A novel geometric approach to binary classification based on scaled convex hulls. IEEE Trans. Neural Netw. 20(7), 1215–1220 (2009)
Liu, Z., Liu, J.G., Chen, Z.: A generalized Gilbert’s algorithm for approximating general SVM classifiers. Neurocomputing 73(1–3), 219–224 (2009)
López, J., Barbero, Á., Dorronsoro, J.R.: An MDM solver for the nearest point problem in Scaled Convex Hulls. In: 2010 International Joint Conference on Neural Networks (IJCNN), pp. 1–8 (2010)
Li, Y., Leng, Q., Fu, Y.: Cross kernel distance minimization for designing support vector machines. Int. J. Mach. Learn. Cybern. 8(5), 1585–1593 (2016). https://doi.org/10.1007/s13042-016-0529-8
Frank, A., Asuncion, A.: UCI machine learning repository. http://archive.ics.uci.edu/ml. Accessed 21 Dec 2021
Chang, C.C., Lin, C.J.: LIBSVM: a library for support vector machines. ACM Trans. Intell. Syst. Technol. (TIST) 2(3), 1–27 (2011)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2022 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Leng, Q., Jiao, E., Liu, Y., Guo, J., Chen, Y. (2022). Cross Distance Minimization for Solving the Nearest Point Problem Based on Scaled Convex Hull. In: Huang, DS., Jo, KH., Jing, J., Premaratne, P., Bevilacqua, V., Hussain, A. (eds) Intelligent Computing Methodologies. ICIC 2022. Lecture Notes in Computer Science(), vol 13395. Springer, Cham. https://doi.org/10.1007/978-3-031-13832-4_17
Download citation
DOI: https://doi.org/10.1007/978-3-031-13832-4_17
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-13831-7
Online ISBN: 978-3-031-13832-4
eBook Packages: Computer ScienceComputer Science (R0)