MFS-MCDM: Multi-label feature selection using multi-criteria decision making

作者：

Highlights：

• We have designed a method for feature selection on the multi-label data.

• We model the feature selection process to a multi-criteria decision-making.

• The TOPSIS method is used to rank the features.

• A subspace learning using Ridge Regression is used to capture the correlation.

• The proposed TOPSIS-based multi-label method outperforms competitive methods.

摘要

•We have designed a method for feature selection on the multi-label data.•We model the feature selection process to a multi-criteria decision-making.•The TOPSIS method is used to rank the features.•A subspace learning using Ridge Regression is used to capture the correlation.•The proposed TOPSIS-based multi-label method outperforms competitive methods.

论文关键词：Information fusion,Multi-criteria decision making,TOPSIS,Entropy,Ridge regression,Multi-label feature selection

论文评审过程：Received 13 May 2020, Revised 11 July 2020, Accepted 3 August 2020, Available online 7 August 2020, Version of Record 19 August 2020.

论文官网地址：https://doi.org/10.1016/j.knosys.2020.106365