On the reliable detection of concept drift from streaming unlabeled data
作者：
Highlights：
• New classifier-independent, dynamic, unsupervised approach for detecting concept drift.
• Reduced number of false alarms and increased relevance of drift detection.
• Results comparable to supervised approaches, which require fully labeled streams.
• Our approach generalizes the notion of margin density, as a signal to detect drifts.
• Experiments on cybersecurity datasets, show efficacy for detecting adversarial drifts.
摘要
•New classifier-independent, dynamic, unsupervised approach for detecting concept drift.•Reduced number of false alarms and increased relevance of drift detection.•Results comparable to supervised approaches, which require fully labeled streams.•Our approach generalizes the notion of margin density, as a signal to detect drifts.•Experiments on cybersecurity datasets, show efficacy for detecting adversarial drifts.
论文关键词：Concept drift,Streaming data,Unlabeled,Margin density,Ensemble,Cybersecurity
论文评审过程：Received 5 December 2016, Revised 15 March 2017, Accepted 3 April 2017, Available online 4 April 2017, Version of Record 8 April 2017.
论文官网地址：https://doi.org/10.1016/j.eswa.2017.04.008
