earticle

영어

In an open and dynamic cloud computing environment, the randomness of cloud services with unreliable quality leads to low accuracy in service composition. To address the above problem, this paper proposes a global trust service composition approach based on random QoS and trust evaluation, considering the multi-criteria assessment of service quality. Firstly, statistical test is employed to remove the uncertain outliers and to estimate the ideal value of the collected objective QoS data. Secondly, subjective QoS evaluations of providers and users are aggregated according to direct trust and recommended trust. Finally, services are composed through global QoS optimization. Experiments for each stage show that the approach improves the accuracy and precision of service composition.