摘要

Cloud computing has been paid more attention due to its outstanding advantages. However, trust issue greatly affects the adoption of cloud services. Selecting trustworthy cloud service from those with same functionality but different qualities has become a significant challenge. Since trustworthiness evaluation is multi-dimensional, how to assign weight for each influence factor is a non-trivial problem. In this paper, we put forward a novel TOPSIS evaluation scheme for cloud service trustworthiness combining objective and subjective aspects. First, we consider the objectivity of cloud services from two facets. For one thing, we concern the reliability of QoS information source and utilize monitored values of QoS attributes rather than feedback ratings from consumers. For another, we employ objective entropy weight for different QoS attributes to decrease the effect of false or artificial parameter information. Second, we introduce trust preference that reflects the subjectivity of trust. Most important of all, we propose the combined weight integrating the two aspects and apply it to TOPSIS method to develop a novel evaluation scheme. The results of two experiments based on the existing QWS dataset from real Web services demonstrate its feasibility, effectiveness, and better Satisfaction Degree from the perspective of consumers.