Sparse Lq-norm least squares support vector machine with feature selection
作者：
Highlights：
• We propose an Lq-norm LS-SVM with feature selection for small size samples.
• Feature selection is achieved effectively by minimizing the Lq-norm of weight.
• The number of selected features can be adjusted by choosing the parameters.
• An efficient iterative global convergent algorithm is introduced to solve the primal problem.
• Experimental results show its feasibility and efficiency.
摘要
•We propose an Lq-norm LS-SVM with feature selection for small size samples.•Feature selection is achieved effectively by minimizing the Lq-norm of weight.•The number of selected features can be adjusted by choosing the parameters.•An efficient iterative global convergent algorithm is introduced to solve the primal problem.•Experimental results show its feasibility and efficiency.
论文关键词：Least squares support vector machine (LS-SVM),Lq-norm,Feature selection,Sparse approximation,Global optimality
论文评审过程：Received 31 October 2016, Revised 11 January 2018, Accepted 22 January 2018, Available online 2 February 2018, Version of Record 2 February 2018.
论文官网地址：https://doi.org/10.1016/j.patcog.2018.01.016
