Fine-Tuning Detection Criteria for Enhancing Anomaly Detection in Time Series

Anomaly detection is the problem of identifying observations that do not conform to the typical ones in a time series. Detection methods implicitly define detection criteria, such as deviation measures, filter thresholds, and candidate anomaly selection strategies. Choosing inappropriate criteria results in inaccurate outputs, generating spurious alerts or missing events. Adjusting these criteria is essential for monitoring systems. To address this challenge, this paper explores the fine-tuning of deviation measures, filter thresholds, and candidate selection strategies. Experimental results show that the proper choice of criteria significantly improves anomaly detection performance, often with greater impact than changing the detection methods.