Evaluating Real-Time Anomaly Detection Algorithms -- The Numenta Anomaly Benchmark

Much of the world's data is streaming, time-series data, where anomalies give significant information in critical situations, examples abound in domains such as finance, IT, security, medical, and energy. Yet detecting anomalies in streaming data is a difficult task, requiring detectors to proc...

Ausführliche Beschreibung

Bibliographische Detailangaben

Veröffentlicht in:	2015 IEEE 14th International Conference on Machine Learning and Applications (ICMLA) S. 38 - 44
Hauptverfasser:	Lavin, Alexander, Ahmad, Subutai
Format:	Tagungsbericht
Sprache:	Englisch
Veröffentlicht:	IEEE 01.12.2015
Schlagworte:	Algorithm design and analysis anomaly detection Benchmark testing benchmarks Detection algorithms Detectors Measurement Real-time systems streaming data time-series data
Online-Zugang:	Volltext
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!