Feature selection approach based on hypothesis-margin and pairwise constraints

In this paper, we propose a semi-supervised margin-based feature selection algorithm called Relief-Sc. It is a modification of the well-known Relief algorithm from its optimization perspective. It utilizes cannot-link constraints only to solve a simple convex problem in a closed form giving a unique solution. Experimental results on well-known datasets validate the effectiveness of our proposed algorithm. Only with little supervision information, Relief-Sc proved to be comparable to supervised feature selection algorithms and was superior to the unsupervised ones.