A new National Science Foundation INTEROP grant has been awarded to UW-Madison through principal investigator D to work on semantic interoperability for geospatial data. The grant is from NSF’s INTEROP program for data interoperability. The main work will be done by the Spatial Ontology Community of Practice (SOCoP) network, which is a core group of active geospatial participants from academia, government, and industry. In the vein of the solicitation, this initial network will be expanded to include others across the country who are interested.
It is all too commonplace that geospatial data from one government agency, researcher, or business enterprise are not able to be combined with data from another agency, etc. even though the data are about the same types of information. Although schema matching or ETL techniques have been developed, along with methods to convert spatial reference systems, semantic issues have not yet been resolved. Examples of semantic mismatches are people using different terms for the same thing, using the same term for different things, or having entirely different conceptual views of a domain.
The INTEROP work has many aspects, such as working to develop formal representations for information in a domain (i.e., create ontologies). There will also be an on-line collaboration environment, work to enhance a repository for geospatial ontologies and tools, educational and participatory workshops for the broad geospatial community to solicit input on use cases, smaller working group sessions on one or two chosen domains to collaborate on developing ontologies that work across use cases, and interactions with standards groups such as OGC. Some work has already been done, for example, with the USGS regarding The National Map version 3.0, which will incorporate intelligence and be semantics-driven. The INTEROP project will also contribute to a model to include semantics into the next phase of the National Spatial Data Infrastructure.