Enabling collaboration and improving understanding: The Management of Semantics for Geospatial Information

Dr. Mark N. Gahegan (Principal Investigator)
Dr. Alan M. MacEachren (Co PI)
Dr. Masa Takatsuka (Co PI)

Where is the wisdom we have lost in knowledge? Where is the knowledge we have lost in information? --T.S. Eliot, The Rock, 1934.

The meaning of geospatial information is constructed, shaped and changed by the interaction of people and systems. But this interaction is not captured, and information becomes separated from the context by which it is given meaning. Information is often constructed incrementally, with successive communities or individuals adding to or modifying meaning through a series of different processes. Those not directly involved in the creation or uncovering of information may not share in this evolving schema fully, neither is its essence conveyed in the existing data models of the systems in which the information resides.

This document proposes a novel research program to establish a semantic framework for managing the meaning associated with geospatial information. The framework will describe the content and various transformations that the geospatial information undergoes, thereby facilitating better understanding, improving communication and promoting best practice in bringing geospatial information to bear. The development of such a framework is described, along with associated rules and knowledge structures to oversee the transformation and interpretation processes. A design for such a framework is outlined. A visual interface to the semantic framework is also proposed, so that its content and meaning are accessible in a form that is readily assimilated. The resulting semantically-enriched geospatial infrastructure will improve data exchange and interoperability between systems, and advance collaboration between human analysts and information users. The framework will associate data with an explicit context that describes its meaning, thereby facilitating its interpretation and application. In short-turning data into information.

Expected major outcomes are as follows:

(1) A multi-perspective account of the needs for, and uses of, semantically-oriented geospatial knowledge, with examples drawn from application domains and real-world problems.

(2) An analysis of the common threads of semantics, conflicts of meaning and schemata, underlying representational models, assumptions and constraints that 'color' the application of geospatial data and can affect understanding.

(3) A prototype framework for geospatial semantic representation that will be represented within a machine as a series of collaborating agents acting together to oversee workflows and information exchange, and also within a visualization environment in a graphical form.

(4) Feedback and collaboration with the OGC and specifically their 'Semantics and Information Community Special Interest Group', thus helping to shape the international research and development agenda in this area.

(5) Specific case studies to establish proof of concept and generality of the framework. A team of highly qualified researchers are included in this proposal, comprising academics from computer science, geographic information science and geography, an expert in geological interoperability and the chief scientist of the Open GIS Consortium (OGC).

Read the abstract to another paper presented at GIScience 2000: The First International Conference on Geographic Information Science. Univerrsity of California Regents, Savannah, GA, Oct. 28-31, 2000. Authors: Gahegan, M., Brodaric, B. and Takatsuka, M.. Designing a framework to support the semantic interoperability of geospatial information.