Framework and knowledge for database integration

• Main Authors: LIM, Ee Peng, CHIANG, Roger Hsiang-Li
• Format: text
• Language: English
• Published: Institutional Knowledge at Singapore Management University 1997
• Subjects: Information technology; Integrated database; Implementation; Firm; Engineering; Methodology; Performance evaluation; Knowledge based; Knowledge structure; Databases and Information Systems
• Online Access: https://ink.library.smu.edu.sg/sis_research/898
• Institution: Singapore Management University

Traditionally, data integration research has focused primarily on understanding integration issues from the data instance and schema perspectives. However, when the integration of heterogeneous databases is performed without considering the semantics of local databases, an incorrectly integrated database may result. Moreover, most integration tasks must be performed manually. In this research, we propose a framework for acquiring the appropriate domain semantics and various types of knowledge needed to detect and reconcile heterogeneities among local databases. By (semi-)automating the acquisition and maintenance of these semantics and knowledge, coupled with an expert system that performs reasoning over these knowledge, database integration processes can be performed at a higher level of automation. This research introduces a database integration framework. The proposed framework provides a foundation to: 1) analyze database integration issues from a broad scope, 2) discuss the impact of database reverse engineering on database integration, 3) distinguish the data warehousing approach from the federated database approach for global query processing and instance integration, and 4) identify various types of knowledge that are required for proper data integration. A systematic classification of these knowledge will facilitate the design of appropriate techniques to acquire and utilize them.