University of Twente Student Theses


Ontology Mapping for Geoinformation Integration

Kipngénoh, Koros Wesley (2010) Ontology Mapping for Geoinformation Integration.

[img] PDF
Abstract:Geoinformation integration provides a basis for manipulation of geospatial datasets across various sources of geodata. It is suitable for unlimited number of applications in the geoinformation domain. The use of conceptual representation of geospatial data and their respective relationships to establish the similarity of concepts and instances in heterogeneous sources, forms part of semantic geoinformation integration. The concept of ontology is the base for semantic data integration. Semantic data in the geoinformation domain is growing. The task of finding a concept in dataset A that correspond to a concept in dataset B, requires mappings between the concepts. This task is gaining importance. Given that semantic datasets are structured using different ontologies, (semi)-automatic ontology mapping techniques need to be utilized before geodata integration and retrieval. Given two ontologies from two different semantic data sources, one must be in a position to tell whether they model the same real world phenomena and measure the degree of semantic similarity. A similarity measurement reasoner is used in the establishment of the links between the compared concepts. The characteristics defining the compared concepts are used by the reasoner in establishing whether they correspond to the same entity. Mappings are defined for comparing two instances and the relations that hold between the compared instances and the compared concepts. These relations between compared instances and concepts are defined using very expressive rules. Similarity measure and the defined threshold are the platforms for defining the mapping relations. The realized mappings are used in an application such as geoinformation retrieval from the knowledge base. This application requires the use of datatype properties and very expressive rules. The whole process from the ontology development, similarity measurement, definition of the mapping relations and utilizing them in an application are discussed in this thesis research. The ontology mapping task requires knowledge modelling using ontologies and reasoning techniques. It also depends on the application context. Thus ensuring deduction of knowledge from large and ambiguous domain specific semantic information sources. Similarity measurement technique can be used to determine which concept and instances represent similar notion. The datatype properties and very expressive rules can be used to integrate and retrieve the information from the knowledge base. Keywords Alignment, GIS, Integration, Land cover, Ontology, Ontology mapping, Ontology matching, Rules, Semantic, Similarity
Item Type:Essay (Master)
Faculty:ITC: Faculty of Geo-information Science and Earth Observation
Programme:Geoinformation Science and Earth Observation MSc (75014)
Link to this item:
Export this item as:BibTeX
HTML Citation
Reference Manager


Repository Staff Only: item control page