English

Real-Time Anomaly Detection for Streaming Analytics

Artificial Intelligence 2016-07-11 v1 Distributed, Parallel, and Cluster Computing Systems and Control

Abstract

Much of the worlds data is streaming, time-series data, where anomalies give significant information in critical situations. Yet detecting anomalies in streaming data is a difficult task, requiring detectors to process data in real-time, and learn while simultaneously making predictions. We present a novel anomaly detection technique based on an on-line sequence memory algorithm called Hierarchical Temporal Memory (HTM). We show results from a live application that detects anomalies in financial metrics in real-time. We also test the algorithm on NAB, a published benchmark for real-time anomaly detection, where our algorithm achieves best-in-class results.

Keywords

Cite

@article{arxiv.1607.02480,
  title  = {Real-Time Anomaly Detection for Streaming Analytics},
  author = {Subutai Ahmad and Scott Purdy},
  journal= {arXiv preprint arXiv:1607.02480},
  year   = {2016}
}
R2 v1 2026-06-22T14:49:35.191Z