No abstract

No abstract

With the continued growth of Service Oriented Architectures (SOA), cross-organizational enterprise collaboration is needed to satisfy complex client needs. To define dependencies between collaborative enterprises and guarantee the required Quality of Service (QoS) at different layers of the SOA, agreements present an effective solution. In fact, using agreements at different layers of SOA gives a clear specification and control of the agreed service levels between different stakeholders involved in the composition. In this context, aggregation functions that allow the management of composite agreements from the atomic ones at the same layer and translation techniques to detect dependencies between agreements at multiple layers are needed. In this paper, we give a survey on intra and inter organizational agreements for enterprise collaborations. We start by presenting the motivation of this paper and give an overview of agreements lifecycle and categories. Then, we give a state of the art of agreements composition approaches. After that, we highlight the techniques used to manage dependencies between agreements composition and we finish this paper by giving a synthetic view of existing works.

No abstract

Automating service selection using semantic approaches have been extensively studied in recent years. In fact, given the big number of provider offers, sourcing of the most relevant service to the client intentions is a complex task especially when providers and customers don't share the same knowledge degree. In particular, differentiating between very similar offers satisfying the same number of client constraints is still a challenging task. In this paper, we present a novel semantic scoring approach that helps clients to select the most appropriate service offer according to their intentions. Our approach detects direct and indirect semantic correspondences between these intentions and the available offers using ontological models. It fairly evaluates these offers and ranks them according to their semantic closeness to the client intentions taking into account both functional and QoS properties. Our ranking is based on a deep examination of provider offers and can distinguish between services that look the same for non expert clients.