A Novel Trustworthiness Measurement Model for Cloud Service

Abstract

Recent surveys show that there is enormous increase of organizations intending to adopt cloud, but one of their major obstructions is the trustworthiness evaluation of cloud service candidates. Performing evaluations of cloud service candidates is expensive and time consuming, especially with the breadth of services available today. In this situation, this paper proposes a novel trustworthiness measurement model to evaluate cloud service trustworthiness. By using the proposed trustworthiness measurement model, we first recognize and categorize trust factors in group of shared factors and unique factors. After special treatment of unique factors, all the trust factors in two groups are located in our designed trust dimension by different weighting and positioning approach. Then all the trust factors are converted to trust vectors, these various trust vectors of the services are considered by the designed multicriteria analysis mechanism which can help us to analyze trust features from different perspectives and provide a comprehensive trustworthiness evaluation. In case the measurement result is inconsistent with user preference, we also provide an adjustment approach based on knowledge learning to enhance the accuracy of measurement result. At the end of this paper, our designed trustworthiness measurement model is validated by several experiments. The experiments are designed based upon real world dataset and the results indicate the accuracy of our measurement can be guaranteed.