Template-based ontology population for smart environments configuration

Abstract

Smart Environments is one of several domains in which Semantic Web technologies are applied nowadays. Ontologies, in particular, are used as core modeling languages for representing devices, systems and environments. Developing such ontologies, that typically involve several device descriptions (individuals) and related information, i.e., individuals of classes contributing to the device model, is often done by a manual, time consuming, and error-prone approach. This paper presents a template based approach, which increases accuracy, ease of use, and time-effectiveness of the ontology population process by reducing the amount of user-given information of about an order of magnitude, with respect to the fully manual approach. User-required information only pertains device features (e.g., name, location, etc.) and never implies knowledge of Semantic Web technologies, thus enabling end-user configuration of smart homes and buildings. Experimental results with a prototypical implementation confirm the viability of the approach on a real-world use case. © Springer International Publishing Switzerland 2014.