Mediation system for dealing with semantic problems in databases
Mohamed Khireddine Kholladi, Aicha Aggoune, Abdelkrim Bouramoul · International Journal of Data Mining Modelling and Management · 2017
Integration of heterogeneous data sources is a difficult task for providing users with the unified interface without semantic heterogeneity problems. We can classify this type of heterogeneity into query-level heterogeneity and data source-level heterogeneity. The first level is related to different expressions defined by different users to the same query. The second level of heterogeneity is related to various representations of the data value and to different definitions to describe the same data. In this context, we propose a novel semantic mediation system to solve the diverse levels of semantic heterogeneity in databases. We use ONTology of Alimentation RISks (ONTARIS), our domain ontology on alimentation risks as well as a shared schema of mediator. The results of the experimental evaluation suggest that correct identification and construction of ontology schema matching play a very important role in solving semantic heterogeneity at both the query and database levels.