Comparison: Discovery on WSMOLX and miAamics/jABC

Abstract

This chapter compares the solutions to the SWS-Challenge discovery problems provided by DERI Galway and the joint solution from the Technical University of Dortmund and University of Postdam. The two approaches are described in depth in Chapters 10 and 13. The discovery scenario raises problems associated with making service discovery an automated process. It requires fine-grained specifications of search requests and service functionality including support for fetching dynamic information during the discovery process (e.g., shipment price). Both teams utilize semantics to describe services, service requests and data models in order to enable search at the required fine-grained level of detail.The DERI solution is based on the Semantic Web services framework described in Chapter 10, including a conceptual model for Semantic Web services (Web Service Modeling Ontology, WSMO[1]), a language for service modeling (Web Service Modeling Language, WSML[2]), a middleware system (Web Service Execution Environment, WSMX[3]) and a modelling framework (Web Service Modelling Toolkit, WSMT5). In order to model the scenario, WSMO has been used for modeling of goals, services (i.e. required and offered capabilities) and ontologies (i.e. information models on which services and goals are defined), all expressed in the WSML ontology language. WSMO Mediators were not utilized since a common domain shipment ontology was used in both the goal and service descriptions.