A Unified Schema Matching Framework

The proliferation of applications dealing with shared data radically increases the need to identify and discover the semantically corresponding elements. To cope with the difficulties of the necessary schema matching, we propose a unified framework. The framework tries to collect the most well-known work concerning schema matching in a generalized approach. We observe that nearly all of this work share in their phases but each phase may has its own steps depending on the nature of the matching approach. Therefore, we suggest a unified approach to give a complete and integrated view about schema matching. Throughout this paper, we take mainstream matching systems, namely COMA, Cupid, SemInt, and LSD as main examples to demonstrate the applicability of our approach. This paper aims mainly at presenting a unified approach for the schema matching problem.