Research On Trustworthiness Evaluation Model Of User Behavior In Cloud Environment

Since the concept of cloud computing has been proposed, it has become a widespread concern in industry and academia. However, due to the security problem of resource sharing, Its rapid development has been restricted. In terms of cloud security, whether cloud service providers can conduct an effective assessment of the user's behavior or not has become an important question, and the research of the trust evaluation model of user behavior still belongs to the prospective theory research topic. Among them, evaluation, prediction and control are three aspects of the trusted management mechanism of user's behavior respectively. As the evaluation work is the basis for subsequent two, so it becomes more and more important to develop an effective and scientific evaluation model of user behavior. In this paper, we mainly study the trusted evaluation of user behavior in cloud computing environment, and the main innovations are as follows:Firstly, research credibility calculation method. In order to make the user's behavior can be decomposed into a number of related attributes, using the AHP analytic hierarchy process, and further to refine behavioral attributes as evidence of behavior, making hardware and software direct measurement or simulation evidence convenient. When calculating the value of evidence to evaluate the user's behavior, takes into account the value of the evidence recorded in the past,those three values are calculated by using the corresponding formula to obtain a more objective evidence value to participate in the calculation of the credibility. And at the time of calculation,fully taking into account the time factor, third party's recommendation value, the reward and punishment factor, make the evaluation result more real and scientific.Secondly, rough set theory is cited, in the calculation process of the credibility of user behavior, by using the method of attribute importance, a variable named wrange is defined according to the network environment. Use the measured value and the past one to do subtraction,if the result is greater than or equal to wrange, then, according to the idea of attribute importance,the value of the evidence is very important, we still use the formula mentioned in the third chapter to calculate the value of evidence. On the contrary, if the result is less than the variable,according to the important degree of attribute, the measured value can be directly used as the final value of the calculation. So the calculation procedure of the evidence is simplified, and improving the efficiency of the calculation of evidence is realized, and then it can improve the overall efficiency of evaluating user's behavior.Finally, to regard the transaction between cloud service providers and users as the game relationship, according to the actual situation of the transaction to calculate the revenue of cloud service providers, and defines a variable named prof to indicate the situation, then use it to do the weighted average calculation with the evaluated value obtained before the transaction to get a new value named Voub, this value that is given by the actual transaction, it is more scientific and reliable. And in order to reflect the dynamic performance of the evaluation results, it also set up a relatively small time coefficient named Tp. As long as the cloud end users are within it to send arequest to the cloud service provider, cloud service providers will be based on the size of the value named Voub, to give the corresponding level of the corresponding services.