English

Precision and Recall for Range-Based Anomaly Detection

Artificial Intelligence 2018-02-13 v3

Abstract

Classical anomaly detection is principally concerned with point-based anomalies, anomalies that occur at a single data point. In this paper, we present a new mathematical model to express range-based anomalies, anomalies that occur over a range (or period) of time.

Keywords

Cite

@article{arxiv.1801.03175,
  title  = {Precision and Recall for Range-Based Anomaly Detection},
  author = {Tae Jun Lee and Justin Gottschlich and Nesime Tatbul and Eric Metcalf and Stan Zdonik},
  journal= {arXiv preprint arXiv:1801.03175},
  year   = {2018}
}
R2 v1 2026-06-22T23:41:01.252Z