Schema Evolution publication categorizer

In this paper, we describe an in-progress effort to augment the data integration capabilities of a service-oriented data integration product with the advanced matching capabilities of a knowledgebased schema matching workbench, thereby providing a guided, interactive, and “what”-oriented design environment for use by data architects and integration engineers. The data integration product is the BEA AquaLogic Data Services Platform, which provides a declarative, XML-based foundation for integrating and service-enabling heterogeneous enterprise data sources in order to deliver data services for use by SOA applications. The schema matching tool is the Harmony system from MITRE, a state-ofthe-art prototype system that maintains a growing knowledge base of matching information and employs a composite matching architecture beneath a rich UI to guide the ranking and decisionmaking processes involved in schema matching. We motivate this work, review the capabilities of each of the systems, and sketch the architectural approach that we are taking to combining the two systems into a synergistic whole.