A Referral-Based QoS Prediction Approach for Service- Based Systems

摘要

During development of service-based systems (SBS), the quality of services (QoS) plays animportant role at helping select more suitable services. There have been several QoS prediction approachesproposed, however, their prediction accuracy is low when there are few historical records in the applicationenvironment. In this paper, we propose a new QoS prediction method based on a virtual platformmethodology. The method first constructs a virtual platform based on Gaussian distribution regardingstability and performance of services. With the platform, a referral-based QoS prediction method has beendeveloped to improve prediction accuracy. The experimental results indicate that our method outperformsprevious approaches, achieving higher prediction accuracy, especially when there are few historical recordsavailable.