Skip to main content
All terms
Foundations

Anomaly Detection

Spotting data points that don't fit the normal pattern, such as fraud or equipment faults.

Definition

Anomaly detection is the task of spotting data points that do not fit the normal pattern — a fraudulent transaction, a failing machine, a network intrusion. Because real anomalies are rare and varied, models often learn what 'normal' looks like and flag anything that deviates from it.