Composition Context for Web Services Selection

Abstract

Often there are several services providing similar functionality, moving the problem of selecting the most suitable to the forefront of interest. In this paper we consider the selection of services in a dynamic environment with changing requirements. In previous work we considered selecting services in isolation, here we present an enhancement to select services in their relation to each other to gain a global optimal solution which nevertheless respects local criteria. Novel contributions are the definition of a composition context and the global multi-criteria optimization mechanism.