关键词: 特征选择, 孪生支持向量机, L1范数, 嵌入式方法

Abstract: A new embedded feature selection method based on linear Twin Support Vector Machine（TWSVM） is proposed. It selects features during classifier construction by introducing a penalty term in the primal formulation of Twin Support Vector Machine. In the solving process, it utilizes alternating iterative optimization method to decompose the problem of solving the model into two sub-problems, namely the standard TWSVM optimization problem and the nonlinear constrained optimization problem about feature weight, and effectively solves the sub-problems respectively. The feature selection method is analyzed and compared on UCI datasets. Simulation results verify the proposed method is effective.

Key words: feature selection, Twin Support Vector Machine（TWSVM）, L1-norm, embedded methods