云计算环境下网格用户行为信任模型研究
Trust model for user behavior in cloud computing environment

在云计算环境中网格用户之间的信任是网格安全的重要基础，为充分利用先验知识，文中运用神经网络理论对信任进行了建模，在相识社区的基础上，完成了推荐信任的计算，并采用RBF神经网络的惩罚项理论解决了恶意推荐问题。仿真实验模拟云环境下网格节点文件下载服务，分别使用文中模型、EigenTrust和NoTrust 3种方法来执行文件下载服务的选择过程，实验证明在云环境下的服务网格，使用文中的信任模型方法，可以有效地评估网格用户可信度，提高用户的服务满意度，为云环境下网格用户的行为信任研究提供了新的思路。
The trust between grid users is an important basis for grid security in the cloud computing environment. A neural network theory is used to model the trust on the basis of the community and complete the recommended trust calculations.This paper uses the penalty terms of RBF neural network theory to solve the problems of malicious recommendation. The cloud mesh node of file download services is simulated by using the model proposed in this paper, EigenTrust and NoTrust to perform the selection process for file download services, respectively. Experimental results prove that the method for the trust model can evaluate the credibility of grid users and improve customer service satisfaction,thus providing a new idea for the behavior trust of grid users in the cloud environment