Synthetic minority oversampling technique for multiclass imbalance problems
作者：
Highlights：
• Synthetic oversampling technique for multiclass imbalance problems is proposed.
• Mechanism of avoiding over generalization is established.
• Regions of minority classes can aggressively enlarge.
• Effective minimization of class overlapping.
• Superior performance than the state of the art over various classifiers.
摘要
•Synthetic oversampling technique for multiclass imbalance problems is proposed.•Mechanism of avoiding over generalization is established.•Regions of minority classes can aggressively enlarge.•Effective minimization of class overlapping.•Superior performance than the state of the art over various classifiers.
论文关键词：Multiclass imbalance problems,Synthetic minority oversampling,Over generalization,Neighbor directions
论文评审过程：Received 11 January 2017, Revised 3 July 2017, Accepted 25 July 2017, Available online 25 July 2017, Version of Record 4 August 2017.
论文官网地址：https://doi.org/10.1016/j.patcog.2017.07.024
