Linking Context to Ontologies

Big Data Era brings two main dimensions, which are heterogeneity and contextual data. The heterogeneity may occur at syntax and semantic levels. Ontologies are largely used to reduce these heterogeneities. Data has limited value if not paired with its context. Usually the internal data of companies are not connected to the rest of data universe including news, weather, user profiles, etc. In Engineering domain, the context is in the mind of engineers. Joining data to its context has become a challenging issue. Some research works attempts to define contextual ontologies. Note that the development of ontologies is a time consuming task and incorporating context in this development contributes in increasing of this time. In this paper, we propose to first model Context and then link it to existing domain ontologies. In the case, where ontologies are used to design applications such as databases, the context becomes a part of the databases. To show this issue, we present a case study of designing Semantic Database (SDB) system, called OntoDB, developed in our Laboratory, in which context can be queried as any entity of the databases.

I. Introduction

The spectacular growth of the use of ontologies [1] by various sort of applications in several domain has resulted in an explosion of the manipulated data. Consequently, in order to store and manage efficiently these big amounts of data, a new generation of databases called Semantic Databases (SDB) has been introduced. These databases store both data and the ontologies that define their meaning in the same repository [2], [3].