Autonomous composition of fuzzy granules in ambient intelligence scenarios

Pervasive and human-centric computing is beginning to be fact: with cell phones, laptops and handhelds, human beings can work pretty much anywhere. Ambient Intelligence (AmI) is a novel human-centric computer discipline based on three emergent technologies: Ubiquitous Computing, Ubiquitous Communication and Intelligent User Interfaces. The integration of aforesaid technologies opens new scenarios for improving the interaction between humans and information technology equipments realizing a human-centric computing environment. Within this aim the deliverable of tasks or services should be achieved through the usage of an invisible network of heterogeneous devices composing dynamic computational-ecosystems capable of satisfying the users requirements. Fuzzy granules, intended as a clump of objects which are drawn together by criteria like indistinguishability, similarity, proximity or functionality, can represent a powerful and, simultaneously, simple paradigm to embed intelligence into a generic AmI ecosystem in order to support people in carrying out their everyday life activities, tasks and rituals in easy and natural way. However, the strong dinamicity and the high complexity characterizing a typical AmI scenario make difficult and expensive to design ad-hoc fuzzy granules. This paper presents a framework exploiting methodologies coming from Semantic Web and Computational Intelligence areas to compose fuzzy granules in autonomous way in order to maximize the users comfort and achieve the hardware transparency and interoperability.