A smart decision support systems based on a fast classifier and a semantic post reasoner
In modern monitoring systems for critical infrastructures protection, there is the need to design automatic decision support systems with high reliability and real-time constraints. These features are usually in contrast as there is a significant trade-off between performance and precision. In this paper, we propose an innovative approach for smart event detection and enriched phenomena comprehension based on ontological and semantic models. In particular, the proposed approach is based on a two-step process made of a fast classifier for real-time alarm raising and an offline post reasoner to refine the final decision and give operators more feelings about the situation assessment. Details about data processing and alarm detection are illustrated and discussed in a practical case study designed for securing subways infrastructures.
Keywords: monitoring systems, semantic-based models, decision support systems, DSSs, sensor data models