Bhattacharyya distance based concept drift detection method for evolving data stream

作者：

Highlights：

• We introduce a new drift detection method using Bhattacharyya distance metric (BDDM).

• BDDM proposes a single-window method for monitoring overtime the mean and variance.

• Experiments on both real and artificial data show encouraging results for BDDM.

• BDDM presented the best balance of false positive and false negative detections.

摘要

•We introduce a new drift detection method using Bhattacharyya distance metric (BDDM).•BDDM proposes a single-window method for monitoring overtime the mean and variance.•Experiments on both real and artificial data show encouraging results for BDDM.•BDDM presented the best balance of false positive and false negative detections.

论文关键词：Concept drift,Bhattacharyya distance,Massive Online Analysis,Naive Bayes classifier,Hoeffding tree classifier

论文评审过程：Received 10 August 2020, Revised 18 March 2021, Accepted 27 May 2021, Available online 1 June 2021, Version of Record 15 June 2021.

论文官网地址：https://doi.org/10.1016/j.eswa.2021.115303