Supervised Learning: Feedback for Improving Alerting

Container Observability provides a facility for users to provide feedback on the correctness of the anomalies detected by the ML system.

When the user determines that an anomaly is missed on a component, such as a false negative, or is detected incorrectly, such as a false positive, the user can set the Anomaly label to provide feedback or supervise the learning. This ensures that Container Observability’s ML system learns the correct behavior for that component if the same condition occurs again and avoids creating a false alert.

Related topics: