An efficient binary Salp Swarm Algorithm with crossover scheme for feature selection problems

作者：

Highlights：

• Two wrapper feature selection approaches using Salp Swarm Algorithm are proposed.

• The crossover operator is utilized in addition to transfer functions to enhance the algorithm.

• The performance is evaluated based on 22 datasets, and compared to five well-known wrapper methods.

摘要

•Two wrapper feature selection approaches using Salp Swarm Algorithm are proposed.•The crossover operator is utilized in addition to transfer functions to enhance the algorithm.•The performance is evaluated based on 22 datasets, and compared to five well-known wrapper methods.

论文关键词：Wrapper feature selection,Salp Swarm Algorithm,Optimization,Classification,Machine Learning,Data Mining,Evolutionary Computation,Swarm Intelligence

论文评审过程：Received 17 January 2018, Revised 31 March 2018, Accepted 3 May 2018, Available online 9 May 2018, Version of Record 26 May 2018.

论文官网地址：https://doi.org/10.1016/j.knosys.2018.05.009