A geographical ontology of objects in the visible domain

This dissertation describes the development of a very large, multi-domain ontology. It addresses the processes involved in developing object-oriented ontological data structures in general and specifically ontological data structures of the geospatial domain. The ontological data structure developed during this research effort---the Visual Objects Taxonomy/Thesaurus (VOTT)---is a compendium of feature classes and concepts within the visual domain that are of interest to geographers, cartographers, geographic information science (GISci) practitioners, geospatial modeling and simulation engineers, and cognitive scientists. This data structure is an object-oriented, knowledge base of geospatial objects in the visual domain---those objects that we see when we look out a picture window; or objects we see when we look out an aircraft window; even those objects that we see when we view an urban setting. The semantic integration process used in this effort was designed to allow the integration of all forms of existing ontological data structures---word-lists, dictionaries, taxonomies, thesauri, and many other forms of ontologies to create a compendium of class concepts. The final products generated during this phase of the research---a broad-based taxonomy and its counterpart thesaurus---provide object-naming consistence for future cartographic, geographic information science, and modeling and simulation production endeavors. An additional by-product of this research effort is a 3-D model library of generic OpenFlight(TM) models of a representative number of the concepts in the geospatial ontology. These models are in the public domain and are freely distributable.