A Visualization Tool for Digital Library Integration
Ananth Raghavan1, Divya Rangarajan1, Rao Shen1, Marcos André Gonçalves2, Naga Srinivas Vemuri1, Weiguo Fan1, Edward A. Fox1
Schema mapping is a challenging problem. It has come to the fore in recent years, especially in the digital library domain for integrating heterogeneous data. Previous studies have approached schema mapping either from algorithmic or visualization perspectives, with few integrating both. With Schema Mapper [1, 2], we demonstrate a semi-automatic tool for schema integration that combines a novel visual interface with an algorithm-based recommendation engine. Schemas are visualized as hyperbolic trees, thus allowing more schema nodes to be displayed at one time. Matches to user selections are recommended, which makes the mapping operation easier and faster. The user has the option to undo mappings, or can choose to ignore recommended matches and create personal mappings.
Once the user decides to save the mappings, an XSLT style sheet with those mappings is created. It can be applied to files conforming to the source schema to transform them to files conforming to the target schema. Schema Mapper also allows editing the target schema, which includes deleting or renaming a node, or adding a sub-tree of the source schema as a child of a node in the target schema. This proves useful in domains like archaeology, wherein the global schema evolves as new collections are being integrated into the Archaeological DL. Although we developed this tool as part of the ETANA-DL  project, it can be easily extended to other applications. Formative evaluation and a preliminary comparison with MapForce  suggest that Schema Mapper may be usefully employed for schema mapping.
© Copyright 2005 Ananth Raghavan, Divya Rangarajan, Rao Shen, Marcos André Gonçalves, Naga Srinivas Vemuri, Weiguo Fan, Edward A. Fox