[an error occurred while processing this directive]

Journal of Systems Engineering and Electronics ›› 2023, Vol. 34 ›› Issue (4): 827-838.doi: 10.23919/JSEE.2023.000103

• • 上一篇    

  

  • 收稿日期:2022-04-13 接受日期:2023-06-15 出版日期:2023-08-18 发布日期:2023-08-28

Robust least squares projection twin SVM and its sparse solution

Shuisheng ZHOU1(), Wenmeng ZHANG1(), Li CHEN2,3,*(), Mingliang XU3()   

  1. 1 School of Mathematics and Statistics, Xidian University, Xi’an 710126, China
    2 School of Physical Education, Zhengzhou University, Zhengzhou 450001, China
    3 School of Computer and Artificial Intelligence, Zhengzhou University, Zhengzhou 450001, China
  • Received:2022-04-13 Accepted:2023-06-15 Online:2023-08-18 Published:2023-08-28
  • Contact: Li CHEN E-mail:sszhou@mail.xidian.edu.cn;3137710140@qq.com;cli@zzu.edu.cn;iexumingliang@zzu.edu.cn
  • About author:
    ZHOU Shuisheng was born in 1972. He received his M.S. degree in applied mathematics and Ph.D. degree in computer science from Xidian University, Xi ’an, China, in 1998 and 2005, respectively. He is currently a professor in the School of Mathematics and Statistics, Xidian University. His current research interests include optimization algorithm and its application, machine learning, pattern recognition, kernel-based learning, and support vector machines. E-mail: sszhou@mail.xidian.edu.cn

    ZHANG Wenmeng was born in 1996. She received her degree in the School of Mathematics and Statistics, Xidian University. Her current research interests include optimization algorithm and its application, machine learning, pattern recognition, kernel-based learning, and support vector machines. E-mail: 3137710140@qq.com

    CHEN Li was born in 1982. She received her Ph.D. degree in School of Mathematics and Statistics, Xidian University, Xi ’an, China and M.S. degree in mathematics from China Agricultural University, Beijing, China, in 2019 and 2009, respectively. She is currently an associate professor in Zhengzhou University. Her current research interests include optimization algorithm and its application, machine learning, pattern recognition, and support vector machines. E-mail: cli@zzu.edu.cn

    XU Mingliang was born in 1981. He received his Ph.D. degree from the State Key Lab of Computer Aided Design and Computer Graphics, Zhejiang University, China. He is a professor in the School of Computer and Artificial Intelligence, Zhengzhou University, China. His current research interests include computer graphics, multimedia and artificial intelligence. E-mail: iexumingliang@zzu.edu.cn
  • Supported by:
    This work was supported by the National Natural Science Foundation of China (61772020;62202433;62172371;62272422;62036010), the Natural Science Foundation of Henan Province (22100002), and the Postdoctoral Research Grant in Henan Province (202103111)

Abstract:

Least squares projection twin support vector machine (LSPTSVM) has faster computing speed than classical least squares support vector machine (LSSVM). However, LSPTSVM is sensitive to outliers and its solution lacks sparsity. Therefore, it is difficult for LSPTSVM to process large-scale datasets with outliers. In this paper, we propose a robust LSPTSVM model (called R-LSPTSVM) by applying truncated least squares loss function. The robustness of R-LSPTSVM is proved from a weighted perspective. Furthermore, we obtain the sparse solution of R-LSPTSVM by using the pivoting Cholesky factorization method in primal space. Finally, the sparse R-LSPTSVM algorithm (SR-LSPTSVM) is proposed. Experimental results show that SR-LSPTSVM is insensitive to outliers and can deal with large-scale datasets fastly.

Key words: outliers, robust least squares projection twin support vector machine (R-LSPTSVM), low-rank approximation, sparse solution