An experiment on the matching and reuse of XML schemas

XML Schema is becoming an indispensable component in developing web applications. With its widespread adoption and its web accessibility, XML Schema reuse is becoming imperative. To support XML Schema reuse, the first step is to develop mechanism to search for relevant XML Schemas over the web. This paper describes a XML Schema matching system that compares two XML Schemas. Our matching system can find accurate matches and scales to large XML Schemas with hundreds of elements. In this system, XML Schemas are modelled as labeled, unordered and rooted trees, and a new tree matching algorithm is developed. Compared with the tree edit-distance algorithm and other schema matching systems, it is faster and more suitable for XML Schema matching.