Ontology-based discovery of geographic information services - An application in disaster management

Finding suitable information in the open and distributed environment of current geographic information web services is a crucial task. Service brokers (or catalogue services) provide searchable repositories of service descriptions but the mechanisms to support the task of service discovery are still insufficient. One of the main challenges is to overcome semantic heterogeneity caused by synonyms and homonyms during keyword-based search in catalogues. This paper presents a practical case study to what extent ontology-based service discovery can solve these semantic heterogeneity problems. To this end, we apply the Bremen University Semantic Translator for Enhanced Retrieval as a service broker. The approach combines ontology-based metadata with an ontology-based search. Based on a scenario of finding geographic information services for estimating potential storm damage in forests, it is shown that through terminological reasoning the request finds an appropriate match in a service on storm hazard classes. However, the approach reveals some limitations in the context of geographic web service discovery, which are discussed at the end. (c) 2005 Elsevier Ltd. All rights reserved.