Evaluating Real-time Anomaly Detection Algorithms - the Numenta Anomaly Benchmark

12 Oct 2015 Alexander Lavin Subutai Ahmad

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 process data in real-time, not batches, and learn while simultaneously making predictions... (read more)

PDF Abstract

Datasets


Introduced in the Paper:

NAB
TASK DATASET MODEL METRIC NAME METRIC VALUE GLOBAL RANK BENCHMARK
Anomaly Detection Numenta Anomaly Benchmark Numenta HTM NAB score 64.7 # 3
Anomaly Detection Numenta Anomaly Benchmark Twitter ADVec v1.0.0 NAB score 47.1 # 5
Anomaly Detection Numenta Anomaly Benchmark Etsy Skyline NAB score 35.7 # 6
Anomaly Detection Numenta Anomaly Benchmark Random NAB score 16.8 # 8

Methods used in the Paper


METHOD TYPE
🤖 No Methods Found Help the community by adding them if they're not listed; e.g. Deep Residual Learning for Image Recognition uses ResNet