ALSR: An adaptive label screening and relearning approach for interval-oriented anomaly detection

作者：

Highlights：

• Interval-oriented method has positive performance for anomaly detection.

• Label screening related to anomaly intervals increases the utilization of labels.

• Extra learning on finer granularity of anomalies helps for more precise detection.

• Feature set combining prediction and statistics suit to multi-types of anomalies.

摘要

•Interval-oriented method has positive performance for anomaly detection.•Label screening related to anomaly intervals increases the utilization of labels.•Extra learning on finer granularity of anomalies helps for more precise detection.•Feature set combining prediction and statistics suit to multi-types of anomalies.

论文关键词：Anomaly detection,Multi-type KPIs,Machine learning,Interval-oriented

论文评审过程：Received 27 January 2019, Revised 28 May 2019, Accepted 14 June 2019, Available online 17 June 2019, Version of Record 22 June 2019.

论文官网地址：https://doi.org/10.1016/j.eswa.2019.06.028