Glossary term
Glossary term
Evaluation and Benchmarks
The process of identifying outliers in a training set.
Contrast with novelty detection.
Created for this library
A payments platform uses outlier detection to flag merchants with transaction patterns far from their normal behavior.
A manufacturing plant uses outlier detection on sensor streams to surface anomalies before they cause downtime.
A cloud security team uses outlier detection on login telemetry to find accounts with unusual access patterns.
Definition source: Google for Developers Machine Learning Glossary | Creative Commons Attribution 4.0 License