Development and Scalability Evaluation of an Ontology-based Context Representation Service

The state of art in context-aware computing has argued that ontologies are the most suitable technique for the modeling phase of the context information life cycle. This paper proposes the Hermes Widget service, which provides an ontology-based context information representation mechanism. The aim is to investigate the impact of the complexity of an ontology-based context modeling in a real-world scenario. A scalability evaluation of the Hermes Widget service is performed using a well-known digital archive of physiological signals and related data for monitoring patients in ICUs. By demonstrating a linear growth of total time for context representation in contrast with an exponential growth of context information input, the experiments show the scalable behavior of the Hermes Widget service. Experiments' results also demonstrate that the ontology-based modeling used in Hermes Widget can be used in an efficient manner by offering several benefits to applications of that context modeling formalism.