We are all witnessing the current data explosion: social media data, clinical data, system data, CRM data, web data, and lately tons of sensor data! With the advent of the Internet of Things, system and monitoring applications are producing humongous amounts of data which undergo evaluation to optimise costs and benefits, predict future events, classify behaviours, implement quality control, and more.
The newest challenge lies in predicting the “unknown”, i.e. an anomaly. An anomaly is an event that is not part of the system past, an event that cannot be found in the system’s historical data. In the case of network data, an anomaly can be an intrusion, in medicine a sudden pathological status, in sales or credit card businesses a fraudulent payment, and, finally, in machinery a mechanical piece breakdown.
The problem here is: how can we predict something we have never seen, an event that is not in the historical data?
This presentation deals with the recognition of early signs of anomalies in the sensor signals monitoring a working rotor. Here, auto-regressive models are trained on historical data describing the rotor normal functioning over time, and therefore making possible the recognition of early signs of disruption.
During this webinar we will show you how!