A Novel Support Vector Regression Algorithm for Processing Experimental Data with Perturbation

DOI: https://doi.org/10.62517/jbdc.202301314

Author(s)

Yuan Lv, Cong Yi*

Affiliation(s)

School of Mechanical Engineering, Xi’an University of Science and Technology, Xi’an, Shaanxi, China *Correspondence Author.

Abstract

Processing experimental data with uncertainty is a huge challenge for robust regression modeling and high-accuracy data forecasting. The objective of the present study is the development of a robust support vector regression model for processing observed data of independent variables containing polyhedral perturbation and achievement of high forecasting precision. Firstly, the conception of the data collection of polyhedral perturbation is given and analyzed. Secondly, a novel robust support vector regression model is constructed by replacing the original independent variables in a linear support vector regression model with the independent variables containing polyhedral perturbation. Thirdly, the novel robust linear support vector regression method is also expanded to the non-linear regression model. Both the two models are validated by the linear and nonlinear numerical regression experiments. Comparison of the experimental results show that the proposed method provide more accurate predictions than the traditional regression method for processing data with polyhedral perturbation.

Keywords

Polyhedral Perturbation; Robust Support Vector Regression; Convex Quadratic Programming; Prediction Accuracy

References

[1]Francisco. Pereira, Tom. Mitchell. Machine learning classifiers and fMRI: A tutorial overview. NEUROIMAGE 2009, 45, s199-s209. 10.1016/j.neuroimage.2008.11.007 [2]Vapnik V. N, Learning hidden information: SVM+. IEEE International Conference on Granular Computing, 2006. [3]Luo Jian. Robust kernel-free support vector regression based on optimal margin distribution. KNOWL-BASED SYST 2022. 109477. 10.1016/j.knosys.2022.109477 [4]Panagopoulos; Orestis P. Relaxed support vector regression. ANN OPER RES 2019. 191-210. 10.1007/s10479-018-2847-6 [5]Alzalg Baha, Khaled Badarneh. An infeasible interior-point algorithm for stochastic second-order cone optimization. Journal of Optimization THEORY AND APPLICATIONS 2019 324-346. 10.1007/s10957-018-1445-8 [6]Dong Guishan, Xuewen Mu. A novel second-order cone programming support vector machine model for binary data classification. Journal of Intelligent & Fuzzy Systems 39.3 2020, 4505-4513. 10.3233/JIFS-200467 [7]Robert. W. Hanks, Jeffery. D. Weir. Robust goal programming using different robustness echelons via norm-based and ellipsoidal uncertainty sets. EUR J OPER RES 2017, 262, 636-646. 10.1016/j.ejor.2017.03.072 [8]Mir. Saman. Pishvaee, Mohamadreza. Fazli. Khalaf. Novel robust fuzzy mathematical programming methods. APPL MATH MODEL 2016, 40, 407-418. 10.1016/j.apm.2015.04.054 [9]Mohammadreza. Chamanbaz, Giuseppe.Notarstefano. Randomized Constraints Consensus for Distributed Robust Linear Programming. IFAC-PapersOnLine 2017, 50, 4973-4978. 10.1109/TCNS.2020.3024483 [10]Milan. Hladík. Robust optimal solutions in interval linear programming with forall-exists quantifiers. EUR J OPER RES 2016, 254, 705-714. 10.1016/j.ejor.2016.04.032 [11]Julio. López, Sebastián. Maldonado. Double regularization methods for robust feature selection and SVM classification via DC programming. INFORMATION SCIENCES 2018, 429, 377-389. 10.1016/j.ins.2017.11.035 [12]Julio. López, Sebastián. Maldonado. Group-penalized feature selection and robust twin SVM classification via second-order cone programming. NEUROCOMPUTING 2017, 235, 112-121. 10.1016/j.neucom.2017.01.005 [13]Julio. López, Sebastián. Maldonado. Multi-class second-order cone programming support vector machines. INFORMATION SCIENCES 2016, 330, 328-341. 10.1016/j.ins.2015.10.016 [14]Maryam. Babazadeh, Amin. Nobakhti. Robust controllability assessment via semi-definite programming. SYST CONTROL LETT 2016, 98, December 2016, 1-7. 10.1016/j.sysconle.2016.10.001 [15]T. C. E. Cheng, Y. Shafransky. An alternative approach for proving the NP-hardness of optimization problems. EUR J OPER RES 2016, 248, 52-58. 10.1016/j.ejor.2015.06.076 [16]Ye J J, Zhou J. Exact formulas for the proximal/regular/limiting normal cone of the second-order cone complementarity set. Mathematical Programming, 2017, 162: 33-50. 10.1007/s10107-016-1027-1 [17]Liu Yulan, Shaohua Pan. Second-order Optimality Conditions for Mathematical Program with Semidefinite Cone Complementarity Constraints and Applications. Set-Valued and Variational Analysis 30.2 2022, 373-395. 10.1007/s11228-021-00587-z [18]Hang, Nguyen TV; Boris S. Mordukhovich, and M. Ebrahim Sarabi. Augmented Lagrangian method for second-order cone programs under second-order sufficiency. Journal of Global Optimization 82.1 2022, 51-81. 10.1007/s10898-021-01068-1 [19]Zhao Qi; Wenhao Fu; Zhongwen Chen. A sensitivity result for quadratic second-order cone programming and its application. Applications of Mathematics 66 2021, 413-436. 10.21136/AM.2020.0278-19 [20]Xu Zhijun; Jing Zhou. A global optimization algorithm for solving linearly constrained quadratic fractional problems. Mathematics 9.22 2021, 2981. 10.3390/math9222981 [21]Mu Xuewen; Yaling Zhang. An alternating directions method of multipliers for convex quadratic second-order cone programming. Pacific Journal of Optimization 14.2 2018, 369-380. [22]Toan; Nguyen Thi; et al. Second-order KKT optimality conditions for multiobjective discrete optimal control problems. Journal of Global Optimization 79 2021, 203-231. 10.1007/s10898-020-00935-7 [23]Ye Yafen, et al. Robust support vector regression with generic quadratic nonconvex ε-insensitive loss. Applied Mathematical Modelling 82 2020, 235-251. 10.1016/j.apm.2020.01.053 [24]Anagha P, S. Balasundaram, Yogendra Meena. On robust twin support vector regression in primal using squared pinball loss. Journal of Intelligent & Fuzzy Systems 35.5 2018, 5231-5239. 10.3233/JIFS-169807 [25]Jalilvand Nejad, Amir. Robust optimization under correlated polyhedral uncertainty set. COMPUT IND ENG 2016. 82-94. 10.1016/j.cie.2015.12.006 [26]Qui; Nguyen Thanh. Stability of generalized equations under nonlinear perturbations. OPTIM LETT 2018. 799-815. 10.1007/s11590-017-1147-4 [27]Bertsimas D, Sim M. The price of robustness. Operations Research, 2004. 52(1): 35-53.