The Novelti Analytics engine has been designed to autonomously learn the behaviors of IoT devices by processing and applying its adaptive learning approach on real-time data being collected from devices. With that autonomous intelligence, Novelti can detect patterns and eventual anomalies and, in turn, notify them in real time. As the system evolves, Novelti adapts its normality models to keep the behavior repository updated, both to continue understanding the asset behavior and to eventually detect new anomalies.
However, from an operational perspective, this is just the beginning. In order to get the complete value from this adaptive monitoring approach, Novelti builds and continuously updates a behavioral report called Metrics.
Novelti Metrics provide a summary of the system’s evolution in terms of the variables collected for the stream, the occurrences and duration of the specific behaviors, and the corresponding error levels provided by the user via feedback. Metrics are updated continuously, so they can be used as an updated dashboard of the system’s status.
Metrics are not just used for reporting. Novelti makes use of these metrics to build normality models of the system, and then use these them for advanced behavioral monitoring and anomaly detection. You can learn more about Novelti Metrics here and try using them by requesting access to the Novelti platform.