An Automatic Method of Data Warehouses Multi-Dimension Modeling for Distributed Information Systems
Lihong Jiang,Junliang Xu,Boyi Xu,Hongming Cai
DOI: https://doi.org/10.1109/cscwd.2011.5960048
2011-01-01
Abstract:Nowadays many companies built enterprise level data warehouses (DW) for decision making support. However explosive data accumulated in distributed databases in company or across companies with the widely use of Computer Supported Cooperative Work in Design (CSCWD) technologies. Therefore it becomes a time costing task for engineers to construct the multidimensional model of data warehouse. This research presents an ontology approach to eliminate data source heterogeneity aiming to design the conceptual structure of data warehouse automatically. The proposed approach includes a domain ontology mete-data model, which consists of data, concept, ontology and resource repositories, to describe the semantic meaning of the data sources. The supply-driven and demand-driven methodologies are combined together to construct the concept model of DW. By supply-driven method, domain ontology system is extracted bottom-up from data sources; on the other hand, by demand-driven method, relationships of the concepts in the ontology system are extended top-down according to the business process. Furthermore, candidate of the data warehouse concept model is derived according to the relationships of the concepts in the ontology system. After discussing the process of data warehouse designing, a case study is given to show how our method is used in clinic domain. The result shows that ontology method could help users designing data warehouse more easily.