Anomalies, defined as outliers or out-of-distribution observations, are particularly important to be detected in data mining because they may indicate data corruption or faulty behavior. Trust in Artificial Intelligence systems depends on this because their reliability relies on inputs
into the training distribution.

Furthermore, anomaly detection plays an essential role in certifying data obtained from sensors or images, as well as in identifying symptoms that can be used to drive diagnosis reasoning and health management. The main achievements of the chair are the following: Christoffel function based anomaly detection and its application to sensor networks [1], dynamic clustering based anomaly detection and its application to radiation hardening of space electronics [2], anomaly detection based on deviation tracking in irregularly sampled and distorted time series and its application to predictive maintenance of robotic arms [3], nonlinear regression based anomaly detection and its application to fault detection in photovoltaic power plants [4], Knowledge extraction through process mining and its application to production monitoring [5], Diagnosis based on decision trees embedding symbolic classification and its foreseen application to 3D printer monitoring [6]. The chair has achieved to integrate model-based and data-based methods in a hybrid Artificial Intelligence framework. All applications have been performed within industrial collaborations