Object (B)logging: Semantic Self-Description for Cyber-Physical Systems

This paper proposes the object (b)logging framework, a novel general approach for the Semantic Web of Things. It allows associating semantic annotations to real-world objects and events as well as triggering complex objects interactions through advanced resource discovery. Machine learning algorithms are combined with non-standard reasoning services in order to produce a rich and meaningful semantic representation of events starting from a low-level statistical analysis of data gathered form sensing devices. The acquired knowledge is exposed to the outside world like in a blog and progressively enriched during the object’s lifetime. The presented paradigm ideally applies to cyber-physical systems, where several mobile heterogeneous micro-devices cooperate to connote and modify appropriately the environment they are dipped in.