G-Forest: An ensemble method for cost-sensitive feature selection in gene expression microarrays

作者：

Highlights：

• G-Forest, a cost-sensitive feature selection for microarray data is proposed.

• G-Forest is multi-criteria hybrid combination of GA and Random Forest.

• G-Forest was compared with three state-of-the-art algorithms.

• G-Forest improved the accuracy up to 14 % and decreased the cost up to 56 %.

• G-Forest substitutes high-cost features with less-cost features.

摘要

•G-Forest, a cost-sensitive feature selection for microarray data is proposed.•G-Forest is multi-criteria hybrid combination of GA and Random Forest.•G-Forest was compared with three state-of-the-art algorithms.•G-Forest improved the accuracy up to 14 % and decreased the cost up to 56 %.•G-Forest substitutes high-cost features with less-cost features.

论文关键词：Feature selection,Cost-sensitive,Genetic algorithm,Random Forest,Microarray Gene expression,Silent diseases’ diagnosis

论文评审过程：Received 11 March 2020, Revised 27 June 2020, Accepted 7 August 2020, Available online 14 August 2020, Version of Record 23 August 2020.

论文官网地址：https://doi.org/10.1016/j.artmed.2020.101941