R-HEFS: Rough set based heterogeneous ensemble feature selection method for medical data classification

作者：

Highlights：

• A rough set based heterogeneous ensemble feature selection method is proposed.

• A feature support count method is also proposed to select a relevant feature.

• Stability and diversity of the base feature selectors are examined.

• For preprocessing kNN imputation and rough set discretization techniques are used.

• Less redundant and high relevant features are selected from medical datasets.

摘要

•A rough set based heterogeneous ensemble feature selection method is proposed.•A feature support count method is also proposed to select a relevant feature.•Stability and diversity of the base feature selectors are examined.•For preprocessing kNN imputation and rough set discretization techniques are used.•Less redundant and high relevant features are selected from medical datasets.

论文关键词：Ensemble feature selection,Stability,Rough set,Medical data,Classification

论文评审过程：Received 4 June 2020, Revised 11 February 2021, Accepted 21 February 2021, Available online 6 March 2021, Version of Record 23 March 2021.

论文官网地址：https://doi.org/10.1016/j.artmed.2021.102049