Abstract
This paper proposes a new spheres-based support vector machine (SSVM) for binary data classification. The proposed SSVM is formulated by clustering the training points according to the similarity between classes, i.e., it constructs two spheres simultaneously by solving a single optimization programming problem, in which each point is as far as possible away from the sphere center of opposite class and its projection value on the directed line segment between the two centers is as far as possible not larger than the corresponding radius. This SSVM has a perfect geometric interpretation for its dual problem. By considering the characteristics of the dual optimization problem of SSVM, an efficient learning algorithm for SSVM, which can be easily extended to other SVM-type classifiers, based on the gradient descent and the clipping strategy is further presented. Computational results on several synthetic as well as benchmark datasets indicate the significant advantages of the SSVM classifier in the computational cost and test accuracy.
Similar content being viewed by others
Notes
Here, we only list the simplified dual problems by discarding the constant items and multiplying the objective functions by 1 − νk, k = 1, 2.
Let \(\mathcal {X}\) be a set with n different points and \(\frac {1}{n}\leq \mu <1\), the RCH of \(\mathcal {X}\), denoted as \(\mathcal {C}_{\mu }(\mathcal {X})\), is defined as
$$\mathcal{C}_{\mu}(\mathcal{X})=\left\{\boldsymbol{x}:~\boldsymbol{x}=\sum\limits_{i}\alpha_{i}\boldsymbol{x}_{i}, \boldsymbol{x}_{i}\in\mathcal{X},\sum\limits_{i}\alpha_{i}=1, 0\leq\alpha_{i}\leq\mu\right\}. $$Available at: http://www.mathworks.com.
References
Boser B, Guyon L, Vapnik VN (1992) A training algorithm for optimal margin classifiers Proceedings of the 5th annual workshop on computational learning theory. ACM Press, Pittsburgh, pp 144–152
Vapnik VN (1995) The natural of statistical learning theory. Springer, New York
Burges CJC (1998) A tutorial on support vector machines for pattern recognition. Data Min Knowl Disc 2 (2):121–167
Vapnik VN (1998) Statistical learning theory. Wiley, New York
Osuna E, Freund R, Girosi F (1997) Training support vector machines: an application to face detection Proceedings of IEEE computer vision and pattern recognition, San Juan, Puerto Rico, pp 130–136
Joachims T, Ndellec C, Rouveriol C (1998) Text categorization with support vector machines: learning with many relevant features European conference on machine learning No. 10, Chemnitz, Germany, pp 137–142
Schölkopf B, Tsuda K, Vert J-P (2004) Kernel methods in computational biology. MIT Press, Cambridge
Lal TN, Schroder M, Hinterberger T, Weston J, Bogdan M, Birbaumer N, Schölkopf B (2004) Support vector channel selection in BCI. IEEE Trans Biomed Eng 51(6):1003–1010
Smola A, Schölkopf B (2000) Sparse greedy matrix approximation for machine learning Proceedings of the 17th international conference on machine learning, Stanford, USA , pp 911–918
Achlioptas D, McSherry F, Schölkopf B (2002) Sampling techniques for kernel methods Advances in neural information processing systems, vol 14. MIT Press, Cambridge
Fine S, Scheinberg K (2002) Efficient SVM training using low-rank kernel representations. J Mach Learn Res 2(2):243–264
Cortes C, Vapnik VN (1995) Support vector networks. Mach Learn 20:273–297
Osuna E, Freund R, Girosi F (1997) An improved training algorithm for support vector machines Proceedings of the IEEE workshop on neural networks for signal processing, Amelia Island, FL, USA, pp 276–285
Platt J (1999) Fast training of support vector machines using sequential minimal optimization Advances in kernel methods–support vector learning. MIT Press, Cambridge, pp 185–208
Joachims T (1999) Making large-scale SVM learning practical Advances in kernel methods: support vector machine. MIT Press, Cambridge, MA, pp 169–184
Tsang IW, Kwok JT, Cheung P-M (2005) Core vector machines: fast SVM training on very large data sets. J Mach Learn Res 6:363–392
Tsang IW, Kocsor A, Kwok JT (2007) Simpler core vector machines with enclosing balls Proceedings of the 24th international conference on machine learning, Corvallis, OR
Fung G, Mangasarian OL (2003) Finite Newton method for Lagrangian support vector machine classification. Neurocomputing 55(1–2):39–55
Pavlov D, Mao J, Dom B (2000) Scaling-up support vector machines using boosting algorithm Proceedings of the international conference on pattern recognition, vol 2. Barcelona, Spain, pp 2219–2222
Collobert R, Bengio S, Bengio Y (2002) A parallel mixture of SVMs for very large scale problems. Neural Comput 14(5):1105–1114
Lee YJ, Mangasarian OL (2001) RSVM: reduced support vector machines Proceedings of the first SIAM international conference on data mining
Mangasarian OL, Wild EW (2006) Multisurface proximal support vector classification via generalized eigenvalues. IEEE Trans Pattern Anal Mach Intell 28(1):69–74
Jayadeva, Khemchandani R, Chandra S (2007) Twin support vector machines for pattern classification. IEEE Trans Pattern Anal Mach Intell 29(5):905–910
Ghorai S, Mukherjee A, Dutta PK (2009) Nonparallel plane proximal classifier. Signal Process 89 (4):510–522
Chen X, Yang J, Ye Q, Liang J (2011) Recursive projection twin support vector machine via within-class variance minimization. Pattern Recogn 44(10–11):2643–2655
Shao Y, Zhang C, Wang X, Deng N (2011) Improvements on twin support vector machines. IEEE Trans Neural Netw 22(6):962–968
Peng X (2011) TPMSVM: a novel twin parametric-margin support vector machine for pattern recognition. Pattern Recogn 44(10-11):2678–2692
Peng X, Xu D (2013) A twin-hypersphere support vector machine classifier and the fast learning algorithm. Inf Sci 221(1):12–27
Schölkopf B, Smola AJ, Williamson RC, Bartlett PL (2000) New support vector algorithms. Neural Comput 12(5):1207–1245
Crisp DJ, Burges CJC (1999) A geometric interpretation of ν-SVM classifiers Advance neural information processing system, vol 12, pp 244–250
Mavroforakis M, Theodoridis S (2006) A geometric approach to support vector machine (SVM) classification. IEEE Trans Neural Netw 17(3):671–682
López J., Barbero Á, Dorronsoro JR (2011) Clipping algorithms for solving the nearest point problem over reduced convex hulls. Pattern Recogn 44(3):607–614
Mercer J (1909) Functions of positive and negative type and the connection with the theory of integal equations. Philos Trans R Soc Lond Ser A 209:415–446
Hsu CW, Lin CJ (2002) A comparison of methods for multiclass support vector machines. IEEE Trans Neural Netw 13:415–425
Cevikalp H, Triggs B, Yavuz HS, Kücük Y, Kücük M, Barkana A (2010) Large margin classifiers based on affine hulls. Neurocomputing 73(16–18):3160–3168
Hao P-Y (2010) New support vector algorithms with parameteric insensitive/margin model. Neural Netw 23 (1):60–73
Peng X, Xu D (2013) Geometric algorithms for parametric-margin ν-support vector machine. Neurocomputing 99:197–205
Shao Y, Wang Z, Chen W, Deng N (2013) A regularization for the projection twin support vector machine. Knowl-Based Syst 37:203–210
Peng X, Chen D, Kong L (2014) A clipping dual coordinate descent algorithm for solving support vector machines submitted
Ripley BD (1996) Pattern recognition and neural networks. Cambridge University Press, Cambridge
Rätsch G (2000) Benchmark repository, datasets available at http://ida.first.fhg.de/projects/bench/benchmarks.htm
Peng X, Xu D (2012) Twin Mahalanobis distance-based support vector machines for pattern recognition. Inf Sci 200:22–37
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interests
We declare that we have no any conflict with other people or organizations that can inappropriately influence our work.
Additional information
This work is supported partly by the National Natural Science Foundation of China (61202156).
Rights and permissions
About this article
Cite this article
Peng, X. A spheres-based support vector machine for pattern classification. Neural Comput & Applic 31 (Suppl 1), 379–396 (2019). https://doi.org/10.1007/s00521-017-3004-x
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00521-017-3004-x