Schema matching is the problem of finding mappings between different structured schemas. When retrieving information from digital libraries with heterogeneous schemas, the query over a target schema has to be transformed into a query over the schema of the digital library (the source schema) it has been submitted to. Schema mappings define the rules for this query transformation. In this paper we address the issue of automatically learning these mappings; furthermore, we evaluate their effectiveness in accessing distributed, heterogeneous digital libraries.
展开▼