Derivation and implementation of a semantic GIS data model informed by principles of cognition

Abstract

The purpose of this research is to develop a new kind of semantic GIS data model that is better able to represent users’ conceptual models of geographic domains than the conventional vector and raster data models. To this end, I look to the principles of cognition, how humans represent geographic information in their minds, to inform the development of this semantic data model. A three-stage methodology for the derivation and implementation of the semantic data model is presented. First, a conceptual framework of geographic cognition is developed. This framework incorporates principles of cognitive categorization and ‘top-down’ and ‘bottom-up’ information processing. Second, by replacing those elements of cognition with their database representation counterparts, often drawing from the object-oriented modeling paradigm, a semantic data model is derived. Finally, a prototype implementation of the semantic data model is presented using the Java programming language and the object-oriented database Poet as a development platform. This implementation utilizes a rule-based approach for representing categorical information and for the extraction of semantic entities from observational data.