Multi-label learning with kernel local label information

作者：

Highlights：

• Label correlations are used to train a model and predict labels simultaneously.

• Use label correlations and local label information to aid label prediction.

• Introduce a kernel function and the constraint to learn label correlations.

• Extensive experiments verify the effectiveness of the proposed method.

摘要

•Label correlations are used to train a model and predict labels simultaneously.•Use label correlations and local label information to aid label prediction.•Introduce a kernel function and the constraint to learn label correlations.•Extensive experiments verify the effectiveness of the proposed method.

论文关键词：Multi-label learning,Local label information,Kernel function,Label correlations

论文评审过程：Received 9 May 2020, Revised 28 June 2022, Accepted 29 June 2022, Available online 6 July 2022, Version of Record 12 July 2022.

论文官网地址：https://doi.org/10.1016/j.eswa.2022.118027