Semantic Mediation in the National Geologic Map Database (US)

David Percy, Stephen Richard, David Soller

Controlled language is the primary challenge in merging heterogeneous databases of geologic information. Each agency produces databases with different schema, and different terminology for describing the objects within. In order to make some progress toward merging these databases using current technology, we have developed software and a workflow that allows for the manual semantic mediation of these geologic map databases. Enthusiastic support from many state agencies (stakeholders and data stewards) has shown that the community supports this approach. Future implementations will move toward a more Artificial Intelligence-based approach, using expert-systems or knowledge-bases to process data based on the training sets we have developed manually.

Subjects: 1.7 Expert Systems; 10. Knowledge Acquisition

Submitted: Feb 13, 2008

This page is copyrighted by AAAI. All rights reserved. Your use of this site constitutes acceptance of all of AAAI's terms and conditions and privacy policy.