A joint paper with the Semantic Technology Institute and the University of Murcia introducing a semantic approach towards implementing energy efficient lifestyles through behavioural change is presented at the main track of the SEMANTiCS 2016 conference held in Leipzig, at September 12-15, 2016. In this paper, UBITECH’s R&D team with their co-authors present a novel semantics-empowered approach for motivating end-users towards the adoption of energy efficient lifestyles, based on recommendations provided through personalised applications and serious games. As a foundation of our approach, we have designed two semantic models to represent energy consumption and behavioural characteristics of consumers. The Energy Efficiency Semantic Model represents energy consumption data collected from a heterogeneous sensor network, while the Behavioural Semantic Model focuses on energy consumption profile of end-users.
As a matter of fact, the Energy Efficiency Semantic Model represents the information from different type of sensors and building infrastructure from an energy efficiency perspective. Such type of sensors refer mainly to energy consumption, production and storage. In addition to the sensor-oriented information, the semantic model represents also entities relevant to the spatial elements. Thus, information on buildings, rooms, floors or even open areas is also included. The design of Energy Efficiency Semantic Model is based on the reuse of concepts in well-known ontologies and extending them when appropriate, in order to fulfill the objectives of the model. Our model borrows concepts from SAREF ontology for Device and Building, the DUL ontology for Agent as well as the SSN ontology for Sensor Observation.
On the other hand, the Behavioural Semantic Model represents interventions aiming behavioural change rather than simple activity recognition. This forms a basis for the information transferred to the personalised applications and serious games. We build the model around User and Recommendation concepts. Since user’s interaction is mostly performed via mobile devices, we reuse user and device modules of mIO! ontology network which itself reuses FOAF concepts. Another high-level concept that is relevant to recommendations is user’s Situation. A situation is an inferred state that represents user’s current context. Moreover, the Recommendation concept on the other hand is the core of the model. A recommendation consolidates a sequence of activities and a target user. Since the users are allowed to give implicit or explicit feedback to recommendations, we also model the Feedback concept in relation to Recommendation. Finally, we include a Message concept as a subtype of recommendationthat allows us to represent persuasive messages targeting a simple behaviour.
