A novel random multi-subspace based ReliefF for feature selection

作者：

Highlights：

• RBEFF fully consider the diversity of random subspaces.

• RBEFF also consider the contribution of samples to features in feature selection.

• The framework of RBEFF algorithm can adaptively select effective features.

• Average reduction ratio of RBEFF is higher than compared methods.

• Experimental results show the superiority of RBEFF in feature selection.

摘要

•RBEFF fully consider the diversity of random subspaces.•RBEFF also consider the contribution of samples to features in feature selection.•The framework of RBEFF algorithm can adaptively select effective features.•Average reduction ratio of RBEFF is higher than compared methods.•Experimental results show the superiority of RBEFF in feature selection.

论文关键词：Feature selection,Random subspace,ReliefF,Classification,k-nearest neighbors

论文评审过程：Received 10 December 2021, Revised 6 July 2022, Accepted 6 July 2022, Available online 11 July 2022, Version of Record 27 July 2022.

论文官网地址：https://doi.org/10.1016/j.knosys.2022.109400