云计算环境下用户行为信任的判断模型研究  

Cloud Computing Environment User Behavior Trust Judgment Model Research

作　　者：刘进[1] 邓维斌[1] 陈家佳[1] 

机构地区：[1]重庆邮电大学经济管理学院,重庆400065

出　　处：《计算机仿真》2016年第11期361-364,共4页Computer Simulation

基　　金：重庆市教委科学技术研究项目(KJ1500416);重庆邮电大学社会科学基金项目(k2014-111)

摘　　要：针对云计算用户增加,给云计算系统带来的压力,使得云计算环境下用户行为的信任度降低的问题,提出基于相对熵的组合赋权法的云计算环境下用户行为信任的判断建模方法。先利用AHP原理计算云环境下用户行为的直接信任度,结合信息熵理论思想,获取用户行为信任判断决策属性的权重及加权系数,并以加权系数为依据获取组合赋权法的用户信任模糊综合判断向量,建立云计算环境下用户行为的信任判断模型。仿真结果表明,改进算法的平均精度为97%,比传统算法提高了38.6%;稳定性为98%,比传统算法提高了44.1%,而误差率仅为0.001%,将户交互成功率提高到90%、恶意行为检测率提高到91%。说明改进的建模方法有效的提高了云计算环境下用户行为信任判断模型建立的准确性,具有良好的应用价值。A trust judgment modeling method of user behavior in cloud computing environment is proposed by using combination weighting method based on relative entropy. Firstly, the AHP theory is used to calculate the direct credibility of user behavior in cloud environment. Then, the decision attribute weight and weighting coefficient of trust judgment for user behavior are obtained based on the information entropy theory. Finally, the fuzzy synthetic judgment vector of user trust in combination weighting method based on the weighting coefficient is obtained, and the judgment model of user behavior trust in cloud computing environment is built. The simulation results show that the average precision of modified method is 97%. It is 38.6% higher than the one in traditional method. The stability is 98% and is 44.1% higher than traditional method. Moreover, the error rate is only 0.001%. The method improves the success rate of user interaction to 90% and detection rate of malicious to 91%. It shows that the modified modeling method improves the judgment modeling accuracy effectively and has better application value.

关 键 词：云计算 用户行为 相对熵 

分 类 号：TP39[自动化与计算机技术—计算机应用技术]