Title: Implementation of virtual sensors and one-class classifiers to detect anomalous events
Author: Esteban Jove, University of A Coruña, Spain
General description: this conference deals with the analysis and implementation of anomaly detection systems based on intelligent techniques. Specifically, two of the most commonly used strategies for this purpose are presented, studied and analysed. The first consists of the virtual sensor development based on a hybrid intelligent model capable of detecting anomalous situations. The second strategy is based on the use of one-class techniques, used to implement anomaly detection classifiers. Therefore, an analysis and a comparison of both strategies are accomplished through different real applications, highlighting the performance of each one.