A measure of semantic similarity between a reference context and a current context

Article type: Research Article

Authors: Guessoum, Djamel^{a; *} | Miraoui, Moeiz^b | Zaguia, Atef^{a; c} | Tadj, Chakib^a

Affiliations: [a] Electrical Engineering Department, École de Technologie Supérieure, Quebec University, Canada | [b] Higher Institute of Applied Science and Technology, University of Gafsa, Tunisia | [c] Taif University, College of Computers and Information Technology, Saudi Arabia

Correspondence: [*] Corresponding author. E-mail: djamel.guessoum.1@ens.etsmtl.ca.

Abstract: Context-aware applications are intended to facilitate the adaptation of services in a pervasive computing system. The semantic similarity between contexts and the application of a semantic similarity measure as a mechanism for service adaptation are topics that have yet to be thoroughly explored in the literature. This study measured semantic similarities between quantitative contextual and categorical variables in the field of pervasive computing. The measure was applied to a current context and to several reference contexts, which were predefined based on a contextual data set. Built on the overlap measure because of its simplicity, the proposed weighted method is easy to implement and can be used to evaluate the actual weight of each contextual variable.

Keywords: Pervasive computing, context, concept, similarity, reference, context

DOI: 10.3233/AIS-160404

Journal: Journal of Ambient Intelligence and Smart Environments, vol. 8, no. 6, pp. 697-707, 2016