A New Two-Level Clustering Approach for Situations Management in Distributed Smart Environments

Achouri Mounir , Alti Adel (1) , Derdour Makhlouf (2) , Sébastien Laborie (2) , Philippe Roose (2)

1 L.R.S.D. - Laboratoire de Réseau et Systèmes Distribués
2 LIUPPA - Laboratoire Informatique de l'Université de Pau et des Pays de l'Adour

In the field of smart environments, many devices and multimedia-oriented connected objects have gained a significant attention in recent years. There is a critical need to offer to users flexible and efficient service selection among a large set of candidates based on their surrounding environments, the user's current needs and situations. This article aims to provide a new Generic Context-aware Profile ONTOlogy (GCxPOnto) for a semantic description of heterogeneous profiles with a situations management facility related to smart environments. Based on the GCxPOnto ontology model, a two-layered architecture was proposed. It includes a local server with a sematic profile modeling and a central Cloud that provides efficient situation management and good scalability. This article is divided into three phases: profile clustering, situation management, and multimedia information dissemination related to smart services. The experimental results show a good response time and an enhanced situational accuracy.

Domaines: Multimédia [cs.MM]; Informatique ubiquitaire; Web

Achouri Mounir, Alti Adel, Derdour Makhlouf, Sébastien Laborie, Philippe Roose. A New Two-Level Clustering Approach for Situations Management in Distributed Smart Environments. International Journal of Ambient Computing and Intelligence, 2019, 10 (2), pp.91-111. ⟨10.4018/IJACI.2019040107⟩. ⟨hal-02077490⟩