Today there is an explosion in the amount of information that is readily available. Due to improvements in communication technologies more historically separate data systems are being connected. However these disparate sources may not be accessible via the same language or interface and may often have different data models. There have been many attempts to integrate the different schemas to provide the end user with a conceptually unified view of the information space. This task involves in some cases tightly binding the separate database, and in other cases, constructing views over the different databases that would be useful to the users. The other solution is to translate the schemas of target databases into a common model that can then be queried. These various integration methods must be enabled while maintaining the independence of the data servers, and without creating copies of their data.
In this thesis, we study and compare five different architectures that help the users gain access to data across different databases, and compare different approaches to the problem of the resolution of heterogeneity. We also describe the design and implementation of a mapping tool developed for a prototype system providing access to multiple information resources. The tool provides the ability to define correspondences between the ontologies and the different information sources.
Date: Friday, July 30, 1999
Time: 10:30 AM
Place: 550-PGH
Faculty, students, and the general public are invited.
Thesis Advisor: Dr. Marek Rusienkiewicz