Skip to Main Content
The least absolute shrinkage and selection operator lasso is a promising feature selection technique. However, it has traditionally not been a focus of research in ensemble classification methods. In this paper, we propose an algorithm for building an ensemble of classifiers in lasso feature subspaces. The algorithm consists of two stages: the first is a lasso based feature subset selection cycle, which tries to find several discriminant feature subspaces; the second is an ensemble based decisional system that intends to preserve the classification performances in case of nonstationary perturbations. Experimental results on the two-class textured image segmentation problem assess the effectiveness of the proposed approach.