On the design of Bayesian principled algorithms for imbalanced classification
作者：
Highlights：
• Combining different (neutral) principled rebalancing techniques is proposed.
• The combination degree and the rebalancing intensity are found by cross validation.
• Extensive experiments support the effectiveness of the proposal.
• Shallow and deep neural networks and ensembles are used in the experiments.
• The database characteristics that reduce combinations performance are detected.
摘要
•Combining different (neutral) principled rebalancing techniques is proposed.•The combination degree and the rebalancing intensity are found by cross validation.•Extensive experiments support the effectiveness of the proposal.•Shallow and deep neural networks and ensembles are used in the experiments.•The database characteristics that reduce combinations performance are detected.
论文关键词：Bregman divergences,Combined techniques,Imbalance,Parameter selection,Principled rebalance
论文评审过程：Received 28 December 2020, Revised 12 February 2021, Accepted 15 March 2021, Available online 17 March 2021, Version of Record 20 March 2021.
论文官网地址：https://doi.org/10.1016/j.knosys.2021.106969
