A hybrid matching method for geospatial services in a composition-oriented environment

Abstract

With the development of Internet and GIS, large volumes of spatial data, powerful computing resources and many spatial data processing functions are published in the form of Web services. Finding suitable geospatial services in the composition-oriented environment is a crucial task. The semantic Web provides a kind of technology to find and compose various service resources automatically through the Web. This paper proposes a hybrid method for the semantic matching of geospatial services. The method includes two parts. Part 1 puts forward a multi-level semantic matching approach, which matches single geospatial service at four levels: classification, input/output, precondition/effect and the quality of service (QoS). This multi-level matching approach makes single service matching quicker and more accurate. Part 2 puts forward a matching algorithm for a geospatial service chain based on the context. The algorithm adopts a trace algorithm, taking account of the effect of the context. It restricts the input/output parameters of the current service by the input/output parameters of service chain, pre-service and sub-service. It matches the atomic service dynamically in a composition-oriented environment, and accurately converts the abstract model of geospatial services into an executable geospatial service chain. A case study of the flood analysis for the Poyang Lake illustrates the effectiveness of our context-based matching method for geospatial services.