A comparative analysis of methodologies for database schema integration
ACM Computing Surveys (CSUR)
Data integration: the teenage years
VLDB '06 Proceedings of the 32nd international conference on Very large data bases
| Hi-index | 0.00 |
Information with various formats are gathered and organised by different parties. Data Integration (DI) aims to eliminate the syntactic, structural and semantic heterogeneities associated with such information [1], combine them, and provide access interfaces to the user. In practice, apart from these heterogeneity issues, other factors also have impact on the design of integrated data resources. Examples of such factors include the users' requirements, domain knowledge of the end-users and data integrators, query capabilities of the available data sources, incomplete information contained in the data sources, etc. This leads data integration a complex and error-prone process [2]. Many DI tools have been designed to (semi-)automatically assist integrators in DI tasks such as similarity matching and mapping generation. However, the quality of the integrated solutions generated is still difficult to determine and control, especially in reflecting users' requirements in aspects such as completeness, consistency, accuracy, minimality and performance of the integrated resources [3].