Conformal Anomaly Detection on Spatio-Temporal Observations with Missing Data
This work addresses anomaly detection for spatio-temporal systems like traffic monitoring, but it is incremental as it builds on conformal prediction with computational improvements.
The authors tackled the problem of unsupervised anomaly detection in spatio-temporal data with missing values by developing ECAD, a distribution-free method that wraps around regression algorithms and sequentially detects anomalies, demonstrating superior performance on traffic flow data.
We develop a distribution-free, unsupervised anomaly detection method called ECAD, which wraps around any regression algorithm and sequentially detects anomalies. Rooted in conformal prediction, ECAD does not require data exchangeability but approximately controls the Type-I error when data are normal. Computationally, it involves no data-splitting and efficiently trains ensemble predictors to increase statistical power. We demonstrate the superior performance of ECAD on detecting anomalous spatio-temporal traffic flow.