Service Selection in Ubiquitous Environments: A Novel Approach Using CBR and Skyline Computing

Ubiquitous environments are characterized by the abundant number of users, their mobility and the heterogeneity of their devices. These characteristics pose a big challenge to service environments and service selection methods must change accordingly. In this paper, we describe a novel approach for service selection in ubiquitous environments. To handle the dynamic nature of these types of environments, we propose a model capturing the changes in Context, Profile and Preferences (CPP model). Case Based Reasoning (CBR) techniques are applied in our approach in order to guarantee efficient search capabilities and invisibility of treatments. A skyline service selection method based on Index algorithm and map-reduce paradigm is proposed ensuring the reduction of the search time. An example scenario is finally presented to explain in further details the proposed approach and to highlight the advantages of the applied techniques.