Privacy-Aware Web Service Composition and Ranking

Service selection is a key issue in the Future Internet, where applications are built by composing services and content offered by different service providers. Most existing service selection schemas only focus on QoS properties of services such as throughput, latency and response time, or on their trust and reputation level. By contrast, the risk of privacy breaches arising from the selection of component services whose privacy policy is not compliant with customers' privacy preferences is largely ignored. In this paper, the authors propose a novel privacy-preserving Web service composition and selection approach which i makes it possible to verify the compliance between users' privacy requirements and providers' privacy policies and ii ranks the composite Web services with respect to the privacy level they offer. The authors illustrate their approach using an eCommerce Web service as an example of service composition. Moreover, the authors present a possible Java-based implementation of the proposed approach and present an extension to WS-Policy standard to specify privacy related assertions.