ONTOLOGY-BASED QUERY OF TWO DUTCH TOPOGRAPHIC DATA SETS: AN EMERGENCY RESPONSE CASE

摘要

The integration of geo-information has been greatly advanced by the development of Spatial Information Infrastructures. However, the semantic interoperability, which is essential for the integration of geo-information in open and distributed environments, still not sufficiently supported. One possible way to deal with the problem is to use ontologies to formally describe the meaning of (spatial) data. The use of ontology languages and search techniques is believed to provide means to solve the semantic interoperability and allow machine-automation of data integration, due to the possibility to define semantics explicitly and represent it in a machine-processable way. This semantic interoperability is expected to facilitate information search and query in time-critical situations (as emergency response). In this paper we verify this hypothesis and present our initial investigations to build and use data ontologies (upper and local). These ontologies describe the data sets (e.g. topographic, utility, cadastre) for general use, independent from the domain of disaster management. They are created and maintained by the respective data providers. The paper presents the ontologies built for two Dutch topographic data sets: GBKN (scale of 1:1000 to 1:2000) and TOP10NL (scale 1:10,000). Both of them are important for disaster management. GBKN is more useful for the field emergency workers, while TOP10NL has fewer details and therefore is more appropriate for decision makers at higher levels. A significant amount of the content from both the data overlaps (with different level of detail), but some information is available only in one of the data sets. In addition to these local data ontologies that are bound to a specific data set, there is a upper data ontology that mediates between the local data ontologies. Using these ontologies, we demonstrate that the user can specify search criteria for spatial information without having to know the underlying data structures of the data sets. The paper elaborates on the developed prototype application, which combines ontology-based query of semantically different data sets, with existing tools for geo-information processing (i.e. GeoTools).