Discovering direct and indirect matches for schema elements

Automating schema matching is challenging. Previous approaches to automating schema matching focus on computing direct element matches between two schemas. Schemas, however rarely match directly. Thus, to complete the task of schema matching, we must also compute indirect element matches. In this paper we present a framework for generating direct as well as many indirect element matches between a source schema and a target schema. Recognizing expected data values associated with schema elements and applying schema-structure heuristics are the key ideas to computing indirect matches. Experiments we have conducted over several real-world application domains show encouraging results, yielding over 90% precision and recall for both direct and indirect element matches.