From Top-Level to Domain Ontologies: Ecosystem Classifications as a Case Study

Vagueness and the trade-off between the classification and delineation of geographic regions – an ontological analysis

“…Aside from the taxonomic hierarchy, this methodology also identifies hierarchical subdivisions of the surface of the Earth (or large i-parts of it). Formally, the notions of taxonomic tree, partonomic tree, and subdivision tree are distinguished (Bittner 2007 Taxonomic, partonomic, and subdivision trees for Bailey's system are discussed in Section 8.3. For more details see Bittner (2007).…”

“…The discussion of Bailey's system extends existing work by Sorokine and Bittner (2005), Sorokine et al (2006), and Bittner (2007) in which the relations between the taxonomic structure and the partonomic structure in Bailey's system are analyzed. However, Sorokine and Bittner (2005) and Sorokine et al (2006) considered only the special case in which every class/universal has exactly one instance.…”

“…However, Sorokine and Bittner (2005) and Sorokine et al (2006) considered only the special case in which every class/universal has exactly one instance. Bittner (2007) formalized partonomic and taxonomic structures of Bailey's system in a general mereology-based theory. This theory is extended in this study by a formalization of the logical interrelations between the partonomic and taxonomic structures based on the notion of homogeneity of geographic regions with respect to certain geographic qualities.…”

A spatio-temporal ontology for geographic information integration

“…Of course there may be different ways of classifying, resulting in different classification trees and corresponding to different sub-universal relations. ( Bittner (2007b) schemes in the ecoregion realm.) Notice that the sub-universal relation will generate a lattice -in which a universal can have more than one immediate super-universal (multiple inheritance) -rather than a tree if either (i) different classification trees are mixed, 1 or (ii) universals are confused with collections and sub-universal relations are confused with sub-collection relations, or (iii) other mistakes are made as for example, those discussed in (Sorokine and Bittner 2005).…”

“…The entire framework can in this way be used to integrate data from distinct though related domains, for example from geology, environmental planing, oil chemistry, and the soil sciences. How to use the presented framework to build domain ontologies is demonstrated in more detail in the context of ecoregion classifications in (Bittner 2007b). Further examples of such use in a variety of biomedical domains are outlined in (Smith et al 2007).…”