Geographical information retrieval system using semantic relationships between multiple layers

  • Authors:

  • Tran Minh Duc;Toshikazu Nishimura

  • Affiliations:

  • Ritsumeikan University, Kusatsu City, Shiga Prefecture, Japan;Ritsumeikan University, Kusatsu City, Shiga Prefecture, Japan

  • Venue:
  • Proceedings of the 10th International Conference on Information Integration and Web-based Applications & Services
  • Year:
  • 2008

Quantified Score

Hi-index 0.00

Visualization

Abstract

Current geographical information retrieval systems provide users with keyword-based searching capabilities. Users select a layer, input keywords which they want to search, then the retrieval system returns the matching features to users through a map interface. In such a system, each layer only serves as an independent representation for a group of geographical features, manages them physically, and no semantics are counted. Therefore, the problems of semantic heterogeneity caused by localized language and the ambiguity of natural language make that approach insufficient to satisfy the high quality retrieval requirement. In this research, we propose a method to add semantic capabilities to these systems by defining the semantic relationships between multiple layers on the map. Ontologies are utilized for describing the terms in the domain; each layer is set to belong to one or more terms in the Ontology, and their semantic relationships such as synonyms, hypernyms, hyponyms, and sibling are defined through the relationships between the terms. The architecture for Ontology-based Geographical Information Retrieval System (OGIRS) is provided, then an Ontology-based City Map Guide (OntoCMG) prototype system based on our architecture is introduced. The experimental result shows the system supporting semantic geographical information retrieval, it can retrieval the maps that match the retrieval terms on semantic, and improve the retrieval quality.