A framework for dynamic classifier selection oriented by the classification problem difficulty
作者：
Highlights：
• Framework for dynamic classifier selection oriented by the classification difficulty.
• Pool generation considering a better coverage of the problem complexity space.
• Classifier selection based on meta-features related to the problem complexity.
摘要
•Framework for dynamic classifier selection oriented by the classification difficulty.•Pool generation considering a better coverage of the problem complexity space.•Classifier selection based on meta-features related to the problem complexity.
论文关键词：Multiple classifier systems,Classifier pool generation,Dynamic classifiers selection,Classification problem difficulty
论文评审过程：Received 28 March 2017, Revised 26 August 2017, Accepted 30 October 2017, Available online 31 October 2017, Version of Record 8 November 2017.
论文官网地址：https://doi.org/10.1016/j.patcog.2017.10.038
