# DQE: A Semantic-Aware Evaluation Metric for Time Series Anomaly Detection

arXiv: https://arxiv.org/abs/2603.06131 · HTML: https://mlsanity.com/paper/2603.06131/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 7 of its 26 references that are in the map.
- First posted on arXiv 6 Mar 2026.

## Builds on (most cited first, 7 of 7)

- [Unsupervised Anomaly Detection via Variational Auto-Encoder for Seasonal KPIs in Web Applications](https://mlsanity.com/paper/1802.03903/) (2018)
- [CATCH: Channel-Aware multivariate Time Series Anomaly Detection via Frequency Patching](https://mlsanity.com/paper/2410.12261/) (2024)
- [Local Evaluation of Time Series Anomaly Detection Algorithms](https://mlsanity.com/paper/2206.13167/) (2022)
- [Precision and Recall for Time Series](https://mlsanity.com/paper/1803.03639/) (2018)
- [PATE: Proximity-Aware Time series anomaly Evaluation](https://mlsanity.com/paper/2405.12096/) (2024)
- [GCAD: Anomaly Detection in Multivariate Time Series from the Perspective of Granger Causality](https://mlsanity.com/paper/2501.13493/) (2025)
- [CyberCScope: Mining Skewed Tensor Streams and Online Anomaly Detection in Cybersecurity Systems](https://mlsanity.com/paper/2503.00871/) (2025)

## Abstract (arXiv)

Time series anomaly detection has achieved remarkable progress in recent years. However, evaluation practices have received comparatively less attention, despite their critical importance. Existing metrics exhibit several limitations: (1) bias toward point-level coverage, (2) insensitivity or inconsistency in near-miss detections, (3) inadequate penalization of false alarms, and (4) inconsistency caused by threshold or threshold-interval selection. These limitations can produce unreliable or counterintuitive results, hindering objective progress. In this work, we revisit the evaluation of time series anomaly detection from the perspective of detection semantics and propose a novel metric for more comprehensive assessment. We first introduce a partitioning strategy grounded in detection semantics, which decomposes the local temporal region of each anomaly into three functionally distinct subregions. Using this partitioning, we evaluate overall detection behavior across events and design finer-grained scoring mechanisms for each subregion, enabling more reliable and interpretable assessment. Through a systematic study of existing metrics, we identify an evaluation bias associated with threshold-interval selection and adopt an approach that aggregates detection qualities across the full threshold spectrum, thereby eliminating evaluation inconsistency. Extensive experiments on synthetic and real-world data demonstrate that our metric provides stable, discriminative, and interpretable evaluation, while achieving robust assessment compared with ten widely used metrics.