Manifold regularized discriminative feature selection for multi-label learning

Jia Zhang, Zhiming Luo, Candong Li, Changen Zhou, Shaozi Li

Abstract: Highlights•Label correlations are incorporated into the framework via manifold regularization.•An embedded multi-label feature selection method is proposed with sparsity.•An optimization algorithm is developed to solve the problem with convexity.•Experiments demonstrate the feasibility and effectiveness of the proposed method.

External IDs:doi:10.1016/j.patcog.2019.06.003