Real-time anomaly detection in elderly behavior with the support of task models

With today's technology, elderly can be supported in living independently in their own homes for a prolonged period of time. Monitoring and analyzing their behavior in order to find possible unusual situation helps to provide the elderly with health warnings at the proper time. Current studies are focusing on the elderly daily activity and the detection of anomalous behaviors aiming to provide the older people with remote support. To this aim, we propose a real-time solution which models the user daily routine using a task model specification and detects relevant contextual events occurred in their life through a context manager. In addition, by a systematic validation through a system that automatically generates wrong sequences of tasks, we show that our algorithm is able to find behavioral deviations from the expected behavior at different times by considering the extended classification of the possible deviations with good accuracy.