Boolean Particle Swarm Optimization with various Evolutionary Population Dynamics approaches for feature selection problems

作者：

Highlights：

• Boolean PSO is used to propose a feature selection method.

• Various EPD mechanisms have integrated with the BPSO to enhance its search power.

• The BPSO-TEPD method has better performance when compared with nine optimizers.

• The BPSO-TEPD perceives the highest accuracy in 16 out of 22 datasets.

摘要

•Boolean PSO is used to propose a feature selection method.•Various EPD mechanisms have integrated with the BPSO to enhance its search power.•The BPSO-TEPD method has better performance when compared with nine optimizers.•The BPSO-TEPD perceives the highest accuracy in 16 out of 22 datasets.

论文关键词：Particle Swarm Optimization (PSO),Evolutionary Population Dynamics (EPD),Natural selection,Feature selection,Metaheuristics

论文评审过程：Received 14 August 2021, Revised 11 December 2021, Accepted 14 January 2022, Available online 4 February 2022, Version of Record 10 February 2022.

论文官网地址：https://doi.org/10.1016/j.eswa.2022.116550