Semi-supervised multi-label feature selection via label correlation analysis with l1-norm graph embedding
作者：
Highlights：
• A semi-supervised multi-label feature selection method leveraging shared information among multiple labels is proposed.
• A l1-norm based graph matrix is imposed to capture a clear underlying manifold structure.
• A l2,1-norm is imposed to select the most representative features.
• An efficient iterative algorithm is proposed to optimize the non-smooth objective function.
摘要
•A semi-supervised multi-label feature selection method leveraging shared information among multiple labels is proposed.•A l1-norm based graph matrix is imposed to capture a clear underlying manifold structure.•A l2,1-norm is imposed to select the most representative features.•An efficient iterative algorithm is proposed to optimize the non-smooth objective function.
论文关键词：Semi-supervised learning,Feature selection,Multi-label learning,Shared-subspace learning
论文评审过程：Received 28 October 2016, Revised 3 March 2017, Accepted 13 May 2017, Available online 22 May 2017, Version of Record 29 May 2017.
论文官网地址：https://doi.org/10.1016/j.imavis.2017.05.004
