Use this URL to cite or link to this record in EThOS: http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.564657
Title: Dwelling on ontology : semantic reasoning over topographic maps
Author: Thomson, Marie-Kristina
Awarding Body: University College London (University of London)
Current Institution: University College London (University of London)
Date of Award: 2009
Availability of Full Text:
Access from EThOS:
Access from Institution:
Abstract:
The thesis builds upon the hypothesis that the spatial arrangement of topographic features, such as buildings, roads and other land cover parcels, indicates how land is used. The aim is to make this kind of high-level semantic information explicit within topographic data. There is an increasing need to share and use data for a wider range of purposes, and to make data more definitive, intelligent and accessible. Unfortunately, we still encounter a gap between low-level data representations and high-level concepts that typify human qualitative spatial reasoning. The thesis adopts an ontological approach to bridge this gap and to derive functional information by using standard reasoning mechanisms offered by logic-based knowledge representation formalisms. It formulates a framework for the processes involved in interpreting land use information from topographic maps. Land use is a high-level abstract concept, but it is also an observable fact intimately tied to geography. By decomposing this relationship, the thesis correlates a one-to-one mapping between high-level conceptualisations established from human knowledge and real world entities represented in the data. Based on a middle-out approach, it develops a conceptual model that incrementally links different levels of detail, and thereby derives coarser, more meaningful descriptions from more detailed ones. The thesis verifies its proposed ideas by implementing an ontology describing the land use ‘residential area’ in the ontology editor Protégé. By asserting knowledge about high-level concepts such as types of dwellings, urban blocks and residential districts as well as individuals that link directly to topographic features stored in the database, the reasoner successfully infers instances of the defined classes. Despite current technological limitations, ontologies are a promising way forward in the manner we handle and integrate geographic data, especially with respect to how humans conceptualise geographic space.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID: uk.bl.ethos.564657  DOI: Not available
Share: