An intelligent-ranking framework for web services selection process

Abstract

This paper introduces an intelligent framework for selecting the best candidate service from the registered and published services within a particular registry through analysing the providers, the consumers and the registry constraints in addition to the business environment requirements. The proposed framework involves three particular phases conforming to the main structure of service-oriented architecture (SOA) which is the provider, consumer and registry. In each phase, an appropriate intelligent technique is deployed for establishing a particular functionality. Specifically, an intelligent agent technique is run at the provider’s side for validating the encapsulated business within the designed WSs, a data mining model is executed at the registry’s side for classifying the registered WSs with respect to their quality of services (QoS) parameters, and a statistical methodology which combines the registry and consumers’ preferences is applied in order to recommend the best WS expected to operate properly with the target business process.