Modeling social contexts for pervasive computing environments

Empowered smart objects have become a part of people everyday activities due to the evolution of pervasive computing technologies. In such scenario context awareness plays a fundamental role in enabling adaptation and contextual services. Up to now a lot of efforts have been done in representing context information such as location, time and users identity. At the same time not much attention has been dedicated to social contexts description and adaptation. While context-aware infrastructures do not consider social networks structures, we believe physical objects should be social-aware in order to interact with users in a more suitable way. This paper presents So Smart, a framework for modeling social context in a pervasive computing environment and for supporting reasoning of a multi-agent system made of smart objects. We propose a ontology-based context model to address social context representation and classification and to allow automated social context reasoning in a multi-agent system.