FIRE-DES++: Enhanced online pruning of base classifiers for dynamic ensemble selection

作者：

Highlights：

• We propose an improved version of the FIRE-DES framework called FIRE-DES++.

• The FIRE-DES++ framework tackles the main drawbacks from its previous version.

• Results show the proposed framework is more robust to imbalanced and noisy datasets.

• The FIRE-DES++ outperforms FIRE-DES and the state-of-the-art en- semble techniques.

摘要

•We propose an improved version of the FIRE-DES framework called FIRE-DES++.•The FIRE-DES++ framework tackles the main drawbacks from its previous version.•Results show the proposed framework is more robust to imbalanced and noisy datasets.•The FIRE-DES++ outperforms FIRE-DES and the state-of-the-art en- semble techniques.

论文关键词：Ensemble of classifiers,Dynamic ensemble selection,Classifier competence,Prototype selection

论文评审过程：Received 11 January 2018, Revised 27 June 2018, Accepted 31 July 2018, Available online 4 August 2018, Version of Record 18 August 2018.

论文官网地址：https://doi.org/10.1016/j.patcog.2018.07.037