Novel Feature Selection Method for Nonlinear Support Vector Regression

The development of sparse techniques presents a major challenge to complex nonlinear high-dimensional data. In this paper, we propose novel feature selection method for support vector regression, called FS-NSVR, which first attempts solve the problem in regression technology field. FS-NSVR preserves representative features selected system due its use matrix original space. is challenging mixed-integer programming that solved efficiently by using an alternate iterative greedy algorithm. Experimental results on three artificial datasets and five real-world confirm effectively selects discards redundant system. outperforms L1-norm least squares Lp-norm both ability efficiency.