云计算环境下海量用户行为信任评估模型

Behavior Trust Evaluation Model for Massive Users under Cloud Computing Environment

在对云计算环境下海量用户行为信任评估建模的研究过程中,采用当前算法进行评估建模时,当实体数据量增大时,计算的复杂度也会随着加大,存在建模误差大的问题。为确保网络资源安全共享,提出了一种改进动态博弈论算法的云计算环境下海量用户行为信任评估建模方法,先以博弈论思想为根本,依据海量用户的历史和当前行为,利用不完全信息动态博弈对云计算环境下的海量用户进行分类,并计算海量用户行为的直接可信度、间接可信度、可信风险系数、行为活跃度、行为奖惩因子等多个决策属性,从不同角度将云计算环境下的用户行为的可信关系量化,综合成海量用户行为整体的可信度进行评估,从而建立了精确的云计算环境下海量用户行为信任评估模型。仿真证明,改进动态博弈论算法的云计算环境下海量用户行为信任评估建模方法,建模精确度高,适应性较强。

A behavior trust evaluation modeling method for massive users under the cloud computing environment＇ is proposed based on the improved dynamic game theory algorithm. According to the history and the current behaviors, the massive users under the cloud computing environment are classified by using the incomplete information dynamic game, and the multiple decision attributes of the massive user behaviors （ direct credibility, indirect credibility, reliable risk factor, behavior activity and behavior factor for rewards and punishments） are calculated, the reliable relationship of user behaviors is quantitated from different perspectives, and the overall comprehensive reliability of massive user behaviors is evaluated, thus the precise trust evaluation model of the massive user behavior under the cloud computing environment is established. Simulation results verify the high precision and good performance of the proposed method.