MFeature: Towards high performance evolutionary tools for feature selection

作者：

Highlights：

• Eight feature selection approaches using enhanced moth-flame algorithm are proposed.

• The ensemble strategy, SA strategy and crossover scheme are employed to boost binary MFO.

• The performance is assessed on 30 datasets, and compared with seven methods.

摘要

•Eight feature selection approaches using enhanced moth-flame algorithm are proposed.•The ensemble strategy, SA strategy and crossover scheme are employed to boost binary MFO.•The performance is assessed on 30 datasets, and compared with seven methods.

论文关键词：Moth-flame optimization algorithm,Crossover,Ensemble,Simulated annealing,Feature selection,Data classification

论文评审过程：Received 27 July 2019, Revised 26 June 2021, Accepted 21 July 2021, Available online 24 July 2021, Version of Record 12 August 2021.

论文官网地址：https://doi.org/10.1016/j.eswa.2021.115655