Manifold regularized discriminative feature selection for multi-label learning

作者：

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.

摘要

•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.

论文关键词：Multi-label learning,Feature selection,Label correlations,Manifold regularization,Optimization objective

论文评审过程：Received 8 October 2018, Revised 10 April 2019, Accepted 4 June 2019, Available online 5 June 2019, Version of Record 17 June 2019.

论文官网地址：https://doi.org/10.1016/j.patcog.2019.06.003