WSML or OWL? A Lesson Learned by Addressing NFP-based Selection of Semantic Web Services

The addition of semantic information to Web Service descriptions enables for automatic discovery, selection and composition of Web services. In this context, WSMO and OWL-S have been the major initiatives aiming at providing Semantic Web Services with appropriate models, languages and tools that enable for an e ective exploitation of semantic annotations. In this paper, we present and discuss the adoption of WSML and OWL, the languages of WSMO and OWL-S, to address the problem of Web service description and selection based on non-functional properties (NFPs). The aspects taken into account deal with the di erences of writing descriptions and matching rules with the two languages, and the performance evaluation of the reasoners to match service requests and service o ers. The comparison is conducted through the formalization of the Policy Centered Meta-model (PCM) and the implementation of the Policy Matchmaker and Ranker (PoliMaR) framework that support either WSML or OWL.