A Structural Approach to Specify Data Transformations in Data Warehouses

Ping-Yu Hsu ； Jia-Long Wang ； Yen-Liang Chen

《資訊管理研究》 1卷2期 (1997/01) Pp. 83-102

https://doi.org/10.6188/JEB.1997.1(2).05

並列摘要

One of the major challenges in designing a data warehouse, is to design the data transformation sub-system, which transfers data from source databases to a target database. The target database usually contains summarized data derived from the various source databases. The transformation requires the system to transfer data at the attribute level. Traditional process specification methods, such as DFD, Functional Decomposition, and Process Dependence Diagram, cannot specify data dependence relationships at this fine granularity. Nor can they specify the process of finding the relationships. To overcome these problems, we propose a method called DID (Data Integration Diagram). DID is a hierarchical graphic representation system that allows users to document dependence relationships between source and target data at the attribute level. With the hierarchical presentation mechanism, DID works smoothly with structural analysis methods, such as top-down, and bottom-up methods.

並列關鍵字

Data Warehouse ； Data Transformation ； System Analysis ； Database ； Data Model