A hybrid feature selection approach based on information theory and dynamic butterfly optimization algorithm for data classification
作者：
Highlights：
• A hybrid feature selection is proposed for selecting relevant attributes.
• Dynamic Butterfly algorithm is used for search space optimization.
• Our aim was balancing the trade-off between exploration and exploitation.
• Twenty datasets and ten algorithms are used in results comparison.
• Experimental results confirmed the superiority of our method.
摘要
•A hybrid feature selection is proposed for selecting relevant attributes.•Dynamic Butterfly algorithm is used for search space optimization.•Our aim was balancing the trade-off between exploration and exploitation.•Twenty datasets and ten algorithms are used in results comparison.•Experimental results confirmed the superiority of our method.
论文关键词：Feature selection,Dynamic butterfly optimization algorithm,Feature interaction maximization,Mutual information,Classification accuracy
论文评审过程：Received 9 April 2021, Revised 10 November 2021, Accepted 27 January 2022, Available online 4 February 2022, Version of Record 22 February 2022.
论文官网地址：https://doi.org/10.1016/j.eswa.2022.116621
