Generic Methodology for Semantic Data Warehouse Design: From Schema Definition to ETL

Actually, any company needs to collaborate with others to improve their performance and productivity. Ontologies can provide a way to promote collaboration between companies. They contribute on reducing the syntax and semantic conflicts that may occur during the collaboration process. Data warehouse technology is a serious candidate for data-sharing architecture that may be employed within the collaborating companies. The spectacular adoption of domain ontologies by several communities facilitates the explosion of semantic databases sources (SDB) that become candidate for building the semantic data warehouses (SDW). This situation motivates us to deeply formalize the structure of semantic sources in order to propose an automatic construction of a semantic data warehouse SDW. In this paper, we first proposed a generic framework for handling semantic sources. Secondly, the generic ETL steps are incorporated to our framework. Our proposal is validated through a case study, considering Oracle SDB, where each source references a global ontology of the Lehigh University Bench Mark.