Anomaly Detection · Schema

Anomaly

A detected anomaly in a time series or multivariate data stream, including the affected metric, timestamp, severity score, and contextual metadata.

Anomaly DetectionArtificial IntelligenceData ScienceFraud DetectionMachine LearningMonitoringObservabilityOutlier DetectionPattern RecognitionSecurityTime Series

Properties

Name	Type	Description
id	string	Unique identifier for the detected anomaly.
metric_name	string	Name of the metric or signal in which the anomaly was detected.
timestamp	string	ISO 8601 timestamp when the anomaly was detected.
value	number	The observed metric value at the time of the anomaly.
expected_value	number	The expected metric value based on historical patterns.
anomaly_score	number	Normalized anomaly severity score between 0 (normal) and 1 (highly anomalous).
severity	string	Categorical severity level of the anomaly.
direction	string	Whether the anomaly is a spike above expected, a dip below, or bidirectional.
algorithm	string	The detection algorithm that identified this anomaly.
status	string	Current status of the anomaly alert.
series_id	string	Identifier of the time series or data stream this anomaly belongs to.
dimensions	object	Key-value pairs providing additional context dimensions for the anomaly (e.g., region, service, host).
related_anomalies	array	List of related anomaly IDs grouped in the same root cause cluster.

View JSON Schema on GitHub