Combined Feature Selection and Classification using DCA

摘要

In this paper we introduce a new method using the zero-norm l{sub}0 for the combined feature selection-supervised classification problem. Discontinuity at the origin for l{sub}0 makes the solution of the corresponding optimization problem difficult. To overcome this drawback we use a robust DC (Difference of Convex functions) programming approach which is a general framework for non-convex continuous optimisation. We consider a continuous approximation to l{sub}0 in an appropriate way such that the resulting problem can be formulated in terms of a DC program. Our DCA (DC Algorithm) requires the solution of one linear program at each iteration. Preliminary computational experiments on some real-world data sets show that the proposed method is promising for the combined feature selection-classification and more efficient than the standard FSV (Feature Selection concaVe) approach.