When you enable anomaly detection for a metric, CloudWatch applies statistical and machine learning algorithms. These algorithms continuously analyze metrics of systems and applications, determine normal baselines, and surface anomalies with minimal user intervention.
Navigate to CloudWatch Metrics
Create an alarm based on anomalous behaviour
Once you have generated a sufficient amount of anomalous traffic (it shouldn’t take much), the CloudWatch Alarm will move to a state of alarm. This is a fairly unrealist scenario because you do not have enough data to provide good training data to the model (as the application has only just been created) and you only have yourself as a user. However, you have used CloudWatch Anomaly Detection, which utilises machine learning to help you detect possible issues based on the number of successful page hits to your website.