From the moment data is connected through agents, it will take 7 days before Eyer will provide anomaly alerts to the end-user. If you at a later stage connect more agents, they will need 7 days of training data before the anomaly detection are enabled.
In the training period before the anomaly alerts are enabled, the machine learning algorithms will use the data to learn different patterns (baselines) for all metrics that are included in the anomaly detection.
Even after the initial 7 days of training is done, the ML will continue to learn on new data for increased accuracy of anomaly detection.
The reasons for 7 days of training are:
One full week of data, to enable daily and weekly cycles.
More training data reduces number of false positives and negatives.
The last week of data (also after the initial 7 days) will have the highest “weight” in the anomaly detection.
Beyond 7 days we also consider the following data patterns (from current) to classify if there is an anomaly:
14 days ago
21 days ago
28 days ago
monthly and yearly cycles.