Situation Awareness Meets Ontologies: A Context Spaces Case Study

AutorAndrey Boytsov, Arkady Zaslavsky, Elif Eryilmaz, and Sahin Albayrak
QuelleThe Ninth International and Interdisciplinary Conference on Modeling and Using Context (CONTEXT'15) 

Efficiency and appeal of pervasive computing systems strongly de-pends on how well and robustly they represent and reason about context and situations. Populating situation search space and inferring situations from con-text which, in turn, is computed from fusing sensor data and observations re-mains a major research challenge. This paper proposes to use ontologies as rep-resentation of domain knowledge to generate situation search space and then match context with already defined situations. To illustrate the feasibility, a context spaces approach is used to represent, generate and reason about situa-tions as abstractions in a multidimensional space. The proposed approach is evaluated and discussed.