An Ontology-Based Conceptual Modeling Method for Data Warehouse

Accurate conceptual model is the foundation of data warehouse building. Studies show that one of the main reasons why a data warehouse project ends up failure is that the modeling of user needs cannot be built up compressively and correctly. Nowadays the way of conceptual modeling has become a central issue in the field of data warehouse research. This article proposes an ontology-based conceptual modeling method for data warehouse, which extending the BWW model according to the special application requirements of data warehouse. The method consists of two parts which are domain modeling and property modeling. Domain modeling describes user needs by a set of theme. Each theme can be depicting by multi-dimension and each dimension can be classified into several classifications. Property modeling, which can solve the heterogeneous problem, is based on BWW ontology, and to represent the identity of dimension, classification, etc, and the relationship between them. This method can formalize the users´ needs into a conceptual model with semantic information, which can effectively associate the problem space with the solution space.