Improving interpolation-based oversampling for imbalanced data learning

作者：

Highlights：

• Presents a position characteristic-aware interpolation oversampling algorithm.

• Overcomes the over constraint, low-efficiency expansion, and over generalization.

• The interpolation creation way is generalized into filling the categorical attributes.

• Superior performance than the state of the art in terms of various assessment metrics.

摘要

•Presents a position characteristic-aware interpolation oversampling algorithm.•Overcomes the over constraint, low-efficiency expansion, and over generalization.•The interpolation creation way is generalized into filling the categorical attributes.•Superior performance than the state of the art in terms of various assessment metrics.

论文关键词：Class imbalance,Oversampling,Imbalanced data,Clustering

论文评审过程：Received 24 June 2018, Revised 27 June 2019, Accepted 29 June 2019, Available online 5 July 2019, Version of Record 18 November 2019.

论文官网地址：https://doi.org/10.1016/j.knosys.2019.06.034