Classifier ensemble reduction using a modified firefly algorithm: An empirical evaluation
作者：
Highlights：
• We propose a FA variant for classifier ensemble reduction.
• It incorporates both accelerated attractiveness and evading strategies.
• The attractiveness search operation is directed by local and global best solutions.
• The evading mechanism leads the search to avoid less optimal regions effectively.
• It outperforms other search methods for ensemble reduction by a significant margin.
摘要
•We propose a FA variant for classifier ensemble reduction.•It incorporates both accelerated attractiveness and evading strategies.•The attractiveness search operation is directed by local and global best solutions.•The evading mechanism leads the search to avoid less optimal regions effectively.•It outperforms other search methods for ensemble reduction by a significant margin.
论文关键词：Ensemble reduction,Classification,Firefly algorithm
论文评审过程：Received 19 May 2017, Revised 30 September 2017, Accepted 1 October 2017, Available online 3 October 2017, Version of Record 5 November 2017.
论文官网地址：https://doi.org/10.1016/j.eswa.2017.10.001
