Anomaly Detection
Automatically detect unusual patterns in your data using ML
Anomaly Detection Alerts
Get an alert if your latest data looks unusual compared to recent trends.
Configuration
- Use historical data of: Training period for the ML model (e.g., 1 day, 1 week, 1 month)
- Repeat evaluation every: How often to check for anomalies (e.g., every 10 minutes)
- Group By behavior: When using Group By with Anomaly Detection, the system analyzes each group independently. Alerts trigger separately for each group that shows anomalous behavior. You can preview anomalies for individual groups using the dropdown selector in the top-right corner of the preview panel.
Was this page helpful?