Abstract
The semi-supervised support vector machine (S3VM) for classification is introduced for dealing with quantities of unlabeled data in the real world. Labeled data are utilized to train the algorithm and then were adapted to classify the unlabeled data. However, this algorithm has several drawbacks, such as the non-smooth term of semi-supervised objective function negatively affects the classification precision. Moreover, it is required to endure heavy burden in solving two quadratic programming problems with inversion matrix operation. To cope with this problem, this article puts forward a novel class of Bézier smooth semi-supervised support vector machines (BS4VMs), based on the approximation property of Bézier function to the non-smooth term. Because of this approximation, a fast quasi-Newton method for solving BS4VMs can be used to decrease the calculating time scale. This new kind of algorithm enhances the generalization and robustness of S3VM for nonlinear case as well. Further, to show how the BS4VMs can be practically implemented, experiments on synthetic, UCI dataset, USPS dataset, and large-scale NDC database are offered. The theoretical analysis and experiments comparisons clearly confirm the superiority of BS4VMs in both classification accuracy and calculating time.
Similar content being viewed by others
Notes
The UCI dataset can be available at https://archive.ics.uci.edu/ml/ datasets.php and https://cs.nyu.edu/~roweis/data.html.
The USPS datasets are available at http://www.cs.nyu.edu/*roweis/data.html.
References
Bennett KP, Demiriz A (1999) Semi-supervised support vector machines. In: Kearns Michael S, Solla Sara A, Cohn David A (eds) Advances in neural information processing systems. MIT Press, London, pp 368–374
Reddy IS, Shevade S, Murty MN et al (2011) A fast quasi-Newton method for semi-supervised SVM. Pattern Recogn 44(10):2305–2313
Chapelle O, Zien A (2005) Semi-supervised classification by low density separation. In: AISTATS 2005—Proceedings of the 10th International Workshop on Artificial Intelligence and Statistics, pp 57–64
Lanquillon C (2000) Learning from labeled and unlabeled documents: a comparative study on semi-supervised text classification. In: Zighed Djamel A, Komorowski Jan, Żytkow Jan (eds) Lecture notes in computer science. Springer, Berlin, pp 490–497
Liu CY, Jiang ZS, Su XX (2019) Detection of human fall using floor vibration and multi-features semi-supervised SVM. Sensors 19(17):3720
Kumar MP, Rajagopal MK (2019) Detecting facial emotions using normalized minimal feature vectors and semi-supervised twin support vector machines classifier. Appl Intell 49:4150–4174
Lang RL, Lu RB, Zhao CQ (2020) Graph-based semi-supervised one class support vector machine for detecting abnormal lung sounds. Appl Math Comput 364:124487
Ju Z, Gu H (2016) Predicting pupylation sites in prokaryotic proteins using semi-supervised self-training support vector machine algorithm. Anal Biochem 507:1–6
**e XJ (2020) Multi-view semi-supervised least squares twin support vector machines with manifold-preserving graph reduction. Int J Mach Learn Cybern 11(11):2489–2499
Mygdalis V, Iosifidis A, Tefas A et al (2018) Semi-supervised subclass support vector data description for image and video classification. Neurocomputing 278:51–61
Liu CY, Gryllias K (2020) A semi-supervised support vector data description-based fault detection method for rolling element bearings based on cyclic spectral analysis. Mech Syst Signal Process 140:106682
Li Z, Tian Y, Li K et al (2017) Reject inference in credit scoring using semi-supervised support vector machines. Expert Syst Appl 74:105–114
Liu YQ, Liu SY, Gu MT (2009) Polynomial smooth classification algorithm of vector machines. Comput Sci (in Chinese) 36(7):179–181
Yang L, Wang L (2013) A class of smooth semi-supervised SVM by difference of convex functions programming and algorithm. Knowl-Based Syst 41:1–7
Zhang XD, Ma JG (2015) A general cubic spline smooth semi-supervised support vector machine. Chin J Eng 37:385–389
Zhang XD, Ma JG, Li AH et al (2015) Quintic spline smooth semi-supervised support vector classification machine. J Syst Eng Electron 26:626–632
Deng N, Tian Y, Zhang C (2012) Support vector machines: optimization based theory, algorithms, and extensions. Chapman and Hall/CRC, London
Bézier P (1968) Renault uses numerical control for car body design and tooling[C]//Paper Sae 680010, Society of Automotive Engineers Congress
Choi JW, Elkaim GH (2008) Bézier curve for trajectory guidance. World Congr Eng Comput Sci WCECS 2173(1):22–24
Mandad M, Campen M (2020) Bézier guarding: precise higher-order meshing of curved 2D domains. ACM Trans Graph 39(4):103–118
Raja SP (2020) Bézier and B-spline curves - a study and its application in wavelet decomposition. Int J Wavelets Multiresolut Inf Process 18(4):2050030
Zhu YF, Xu G, Ling CN (2019) Construction of energy-minimizing Bézier surfaces interpolating given diagonal curves. J Image Graph 24(11):1998–2008
Wu Q, Wang E (2015) Bézier function smooth support vector regression. ICIC express letters. Part B Appl Int J Res Surv 6:1773–1779
Nocedal J, Wright SJ (1999) Numerical optimization. Springer, New York
Yu J, Vishwanathau SVN, Gunter S et al (2010) A Quasi-Newton approach to nonsmooth convex optimization problems in machine learning. J Mach Learn Res 11:1145–1200
Chen W, Shao Y, Hong N (2014) Laplacian smooth twin support vector machine for semi-supervised classification. Int J Mach Learn Cybern 5:459–468
Ho TK, and Kleinberg EM (1996) “Checkerboard dataset”, http://www.cs.wisc.edu/~musicant/data/ndc/ , accessed on July 20 2020
Demsar J (2006) Statistical comparisons of classifiers over multiple data sets. J Mach Learn Res 7:1–30
Acknowledgements
This work was supported by the Social Science Foundation of China under Grant (18ZDA027).
Author information
Authors and Affiliations
Corresponding authors
Ethics declarations
Conflict of interest
The author declares that they have no conflict of interest.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Wang, E., Wang, ZY. & Wu, Q. One novel class of Bézier smooth semi-supervised support vector machines for classification. Neural Comput & Applic 33, 9975–9991 (2021). https://doi.org/10.1007/s00521-021-05765-6
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00521-021-05765-6