Abstract
Recently, semi-supervised learning (SSL) has attracted a great deal of attention in the machine learning community. Under SSL, large amounts of unlabeled data are used to assist the learning procedure to construct a more reasonable classifier. In this paper, we propose a novel manifold proximal support vector machine (MPSVM) for semi-supervised classification. By introducing discriminant information in the manifold regularization (MR), MPSVM not only introduces MR terms to capture as much geometric information as possible from inside the data, but also utilizes the maximum distance criterion to characterize the discrepancy between different classes, leading to the solution of a pair of eigenvalue problems. In addition, an efficient particle swarm optimization (PSO)-based model selection approach is suggested for MPSVM. Experimental results on several artificial as well as real-world datasets demonstrate that MPSVM obtains significantly better performance than supervised GEPSVM, and achieves comparable or better performance than LapSVM and LapTSVM, with better learning efficiency.
Similar content being viewed by others
Notes
We use b 1 and b 2 instead of −γ 1 and −γ 2 in the original paper [9] only for the unified notation.
According to [10, 11], using the “difference” instead of the “ratio” does not change the geometrical interpretation of GEPSVM, results in standard eigenvalue problems, which are more efficient than the general eigenvalue problems solved in GEPSVM. Moreover, comprehensive comparisons in [10, 11] show that the “difference” has comparable or better performance compared to the “ratio” (GEPSVM), but with the less learning time.
A particle \(\boldsymbol{x}_{i}^{t}\) with higher classification accuracy produces a better fitness value (lower training error). That is, better fitness is represented by lower value.
Matlab code is available at http://www.optimal-group.org/Resource/MPSVM.html.
Classification accuracy is defined as: \(\mathit{Acc} =\frac{\mathrm{TP} + \mathrm{TN}}{\mathrm{TP} + \mathrm{FP} +\mathrm{TN} + \mathrm{FN}}\), where TP, TN, FP and FN are the number of true positive, true negative, false positive and false negative, respectively.
We use the training time T train and parameter search time T para to denote the computational efficiency (learning time) for each algorithm.
Matlab is available at http://www.mathworks.com.
The UCI datasets are available at http://archive.ics.uci.edu/ml.
The USPS datasets are available at www.cs.nyu.edu/~roweis/data.html.
References
Vapnik VN (1998) Statistical learning theory. Wiley, New York
Burges CJC (1998) A tutorial on support vector machines for pattern recognition. Data Min Knowl Discov 2(2):121–167
Deng N, Tian Y, Zhang C (2013) Support vector machines: theory, algorithms and extensions. CRC Press, Philadelphia
Hao P, Chiang J, Lin Y (2009) A new maximal-margin spherical-structured multi-class support vector machine. Appl Intell 30(2):98–111
Zhang HH, Ahn J, Lin XD, Park C (2006) Gene selection using support vector machines with non-convex penalty. Bioinformatics 22(1):88–95
Lee L, Wan C, Rajkumar R, Isa D (2012) An enhanced support vector machine classification framework by using Euclidean distance function for text document categorization. Appl Intell 37(1):80–99
Lee L, Rajkumar R, Isa D (2012) Automatic folder allocation system using Bayesian-support vector machines hybrid classification approach. Appl Intell 36(2):295–307
Wang C, You W (2013) Boosting-SVM: effective learning with reduced data dimension. Appl Intell 39(3):465–474
Mangasarian OL, Wild EW (2006) Multisurface proximal support vector machine classification via generalized eigenvalues. IEEE Trans Pattern Anal Mach Intell 28(1):69–74
Shao Y, Deng N, Chen W, Zhen W (2013) Improved generalized eigenvalue proximal support vector machine. IEEE Signal Process Lett 20(3):213–216
Ye Q, Zhao C, Zhang H, Ye N (2011) Distance difference and linear programming nonparallel plane classifier. Expert Syst Appl 38(8):9425–9433
Jayadeva KR, Chandra S (2007) Twin support vector machines for pattern classification. IEEE Trans Pattern Anal Mach Intell 29(5):905–910
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 Recognit 44(10–11):2678–2692
Qi Z, Tian Y, Shi Y (2013) Structural twin support vector machine for classification. Knowl-Based Syst 43:74–81
Shao Y, Deng N, Yang Z, Chen W, Wang Z (2012) Probabilistic outputs for twin support vector machines. Knowl-Based Syst 33:145–151
Shao Y, Deng N, Yang Z (2012) Least squares recursive projection twin support vector machine for classification. Pattern Recognit 45(6):2299–2307
Qi Z, Tian Y, Shi Y (2012) Twin support vector machine with universum data. Neural Netw 36:112–119
Qi Z, Tian Y, Shi Y (2013) Robust twin support vector machine for pattern classification. Pattern Recognit 46(1):305–316
Ding S, Yu J, Qi B, Huang H (2013) An overview on twin support vector machines. Artif Intell Rev. doi:10.1007/s10462-012-9336-0
Chapelle O, Schölkopf B, Zien A (2010) Semi-supervised learning. MIT Press, Massachusetts
Zhu X, Goldberg AB (2009) Introduction to semi-supervised learning. Morgan & Claypool, San Rafael
Tur G, Hakkani D, Schapire RE (2005) Combining active and semi-supervised learning for spoken language understanding. Speech Commun 45(2):171–186
Guzella TS, Caminhas WM (2009) A review of machine learning approaches to spam filtering. Expert Syst Appl 36(7):10206–10222
Zhang T, Liu S, Xu C, Lu H (2011) Boosted multi-class semi-supervised learning for human action recognition. Pattern Recognit 44(10–11):2334–2342
Nguyen T, Ho T (2012) Detecting disease genes based on semi-supervised learning and protein protein interaction networks. Artif Intell Med 54(1):63–71
Soares RGF, Chen H, Yao X (2012) Semisupervised classification with cluster regularization. IEEE Trans Neural Netw Learn Syst 23(11):1779–1792
Fan M, Gu N, Qiao H, Zhang B (2011) Sparse regularization for semi-supervised classification. Pattern Recognit 44(8):1777–1784
Belkin M, Niyogi P, Sindhwani V (2006) Manifold regularization: a geometric framework for learning from labeled and unlabeled examples. J Mach Learn Res 7:2399–2434
Melacci S, Belkin M (2011) Laplacian support vector machines trained in the primal. J Mach Learn Res 12:1149–1184
Qi Z, Tian Y, Shi Y (2012) Laplacian twin support vector machine for semi-supervised classification. Neural Netw 35:46–53
Chen W, Shao Y, Ye Y (2013) Improving Lap-TSVM with successive overrelaxation and differential evolution. Proc Comput Sci 17:33–40
Chen W, Shao Y, Hong N (2013) Laplacian smooth twin support vector machine for semi-supervised classification. Int J Mach Learn Res Cybern. doi:10.1007/s13042-013-0183-3
Tikhonov AN, Arsenin VY (1979) Methods for solving ill-posed problems. Nauka, Moscow
Parlett B (1998) The symmetric eigenvalue problem. SIAM, Philadelphia
Lin SW, Ying KC, Chen SC, Lee ZJ (2008) Particle swarm optimization for parameter determination and feature selection of support vector machines. Expert Syst Appl 35(4):1817–1824
Shao Y, Wang Z, Chen W, Deng N (2013) Least squares twin parametric-margin support vector machine for classification. Appl Intell 39(3):451–464
Huang CL, Dun JF (2008) A distributed pso-svm hybrid system with feature selection and parameter optimization. Appl Soft Comput 8(4):1381–1391
Das S, Suganthan PN (2011) Differential evolution: a survey of the state-of-the-art. IEEE Trans Evol Comput 15(1):4–31
Kennedy J, Eberhart R (1995) Particle swarm optimization. In: IEEE international conference on neural networks, vol 4, pp 1942–1948
Poli R, Kennedy J, Blackwell T (2007) Particle swarm optimization. Swarm Intell 1(1):33–57
Gan H, Sang N, Huang R, Tong X, Dan Z (2013) Using clustering analysis to improve semi-supervised classification. Neurocomputing 101:290–298
Yang Z, Fang K, Kotz S (2007) On the student’s t-distribution and the t-statistic. J Multivar Anal 98(6):1293–1304
Acknowledgements
The authors would like to thank the editors and the anonymous reviewers, whose invaluable comments helped improve the presentation of this paper substantially. This work is supported by the National Natural Science Foundation of China (11201426, 61203133, 11301485 and 61304125), the Zhejiang Provincial Natural Science Foundation of China (LQ12A01020, LQ13F030010) and the Science and Technology Foundation of Department of Education of Zhejiang Province (Y201225179).
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Chen, WJ., Shao, YH., Xu, DK. et al. Manifold proximal support vector machine for semi-supervised classification. Appl Intell 40, 623–638 (2014). https://doi.org/10.1007/s10489-013-0491-z
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10489-013-0491-z