Applying a knowledge based system for metadata integration for data warehouses

Abstract

Data warehouses is a typical example of distributed systems where diverse tools and platforms need to communicate to understand each other. For the communication, metadata integration is significant. Seamless metadata interchange improves the data quality and the system effectiveness. Metadata standards exist, for instance, Common Warehouse MetaModel (CWM), which have enhanced the metadata integration. However, it is far from solving the problem of metadata integration in data warehouse environment. This paper proposes an approach to apply a knowledge-based system that supports the metadata integration. By utilizing the knowledge of software engineers on Common Warehouse MetaModels and the metadata interchange models, the knowledge-based system can give metadata interchange model suggestions. Such a knowledge-based system intends to partly automate the metadata integration to improve the efficiency and the quality of metadata integration in data warehouses. © Springer-Verlag 2010.