具有抗攻击能力的信誉与信任模型  

Attack-resistant model of reputation and trust

作　　者：郑小林[1] 刘驰[1] 贝毅君[1] 陈德人[1] 

机构地区：[1]浙江大学计算机科学与技术学院,浙江杭州310027

出　　处：《浙江大学学报（工学版）》2011年第3期405-411,共7页Journal of Zhejiang University：Engineering Science

基　　金：国家自然科学基金资助项目(61003254);国家科技支撑计划资助项目(2008BAH24B03);中国博士后科学基金资助项目(20090451446)

摘　　要：针对互联网第三方交易平台中的信誉体系容易受到欺诈行为攻击,以及不能真实反映买卖双方信任度的问题,提出一种基于社会网络的具有抗攻击能力的信誉与信任模型.基于交易者评分建立信誉评估算法,通过信任链传递算法建立基于社会网络的信任评估模型,并采用二维聚类过滤算法实现对欺诈数据的排查.通过引入纠偏参数,实现信誉值的客观化与信任值的主观化,为交易者做出最终交易提供决策帮助.实验表明,该信誉与信任模型能够客观反映交易者的服务质量以及交易双方的信任关系,并对不诚实反馈和串通欺骗等攻击行为具有较高的抵抗能力.The reputation mechanism of the third party trading platform is easy to be attacked by fraud,and it always cannot characterize the actual trustworthy between the buyers and sellers.To solve this problem,an attack-resistant reputation and trust model based on social network was proposed.A reputation model was firstly established based on dealer＇s rating,then trust was modeled using the transfer algorithm in a social network constructed by dealing relationships.A correcting parameter was introduced to obtain objective reputation value and subjective trust value,which would help dealers make decisions.In addition,two-dimensional cluster filtering was employ to get rid of cheating data.The experimental results show that the model can properly describe the trust relationships between dealers,and performs well in resisting attack behaviors such as cheating feedback and shilling attack.

关 键 词：互联网信誉 信任链 社会网络 串通欺骗 抗攻击 

分 类 号：TP319[自动化与计算机技术—计算机软件与理论] TP391[自动化与计算机技术—计算机科学与技术]