Building an Architecture for Discovery, Selection, Invocation and Personalization of Semantic Web Services

We propose a two-step matchmaking procedure for a Web Service oriented architecture to cope with scalability problems if a large amount of Web Services has to be processed. We distinguish between domain-aware and domainindependent matchmaking, and show how such a two-step matchmaking process can be realized. We discuss the approach within the Personal Reader architecture which enables the use of Web Service based applications to personalize Semantic Web content.