Ontology matching for spatial data retrieval from Internet portals

Abstract. This paper analyses reported difficulties in spatial data retrieval from Internet portals, based on results from an empirical survey. The analysis reveals the problems which cause user dissatisfaction and failure in the search process for spatial data sets. These problems are addressed in a new data search architecture. Within the architecture, the paper focuses on query expansion, which helps to overcome mismatches between the user's taxonomy of geographic features and the taxonomies in the knowledge base of the metadata search engine. The proposed data search architecture is formalized in an algebraic specification language, followed by the simulation of a search scenario for spatial data. The simulation demonstrates the advantage of quer