Example-dependent cost-sensitive adaptive boosting

作者：

Highlights：

• Three variants of AdaBoost adaptation to example-dependent cost-sensitive classification problem.

• Impact of posterior probability calibration on model performance.

• Impact of cost distribution on the model performance.

• Achieving superior classification results comparing to state-of-art methods.

摘要

•Three variants of AdaBoost adaptation to example-dependent cost-sensitive classification problem.•Impact of posterior probability calibration on model performance.•Impact of cost distribution on the model performance.•Achieving superior classification results comparing to state-of-art methods.

论文关键词：Cost-sensitive learning,Example-dependent cost-sensitive classifier,Classifier calibration,Adaptive boosting,AdaBoost,Fraud detection

论文评审过程：Received 13 August 2018, Revised 17 May 2019, Accepted 5 June 2019, Available online 6 June 2019, Version of Record 14 June 2019.

论文官网地址：https://doi.org/10.1016/j.eswa.2019.06.009