一种基于社会网络的激励兼容的信任模型  

Incentive-compatible trust model based on social network

作　　者：罗柏发[1] 蔡国永[1] 

机构地区：[1]桂林电子科技大学计算机科学与工程学院,广西桂林541004

出　　处：《计算机应用研究》2011年第5期1932-1936,共5页Application Research of Computers

基　　金：国家自然科学基金资助项目(61063039);广西研究生教育创新计划资助项目(2009105950812M18)

摘　　要：社会网络中的信任关系是最复杂的社会关系之一,涉及多种因素,很难准确量化和预测。综合考虑影响信任量化的各种因素,在社会网络环境下构建了一个多维决策信任模型。该信任模型引入了直接信任、间接信任、风险函数等关键属性从多个角度动态描述了信任关系的不确定性和复杂性。为抵制恶意实体的非理性行为,提高了信任模型的可信性,在该信任模型中嵌入一个经济激励机制鼓励交互实体诚实合作并激励它们积极参与,最后用密码机制保证激励机制中的安全属性,提高该模型在实际应用中的可信性。仿真实验表明,该信任模型能够准确预测节点的行为信任并控制交互风险,该激励机制能很好地抑制节点的恶意行为,激励信息分享并确保交互过程中敏感数据的安全可信性。The trust relation in the social networks is one of the most complex social relations.It is difficult to be predicted and quantized as it involves too many kinds of factors.Through considering many possible factors comprehensively from the viewpoint of social network,this paper proposed a multi-dimensions trust model,which brcught together direct-trust,indirect-trust,and risk function to describe the complexity and uncertainty of the trust relationship.In order to resist some bad entities＇ fallacious behaviors,designed an economical incentive mechanism in the trust model.The entities would be encouraged to collaborate honestly and actively using this mechanism.Finally,applied a cryptogram mechanism to protect securitive information in the incentive mechanism.And improved the credibility of the model in practical application.Experimental results show that the trust model can not only predicate the behavioral trust of the nodes accurately,but also keep the interaction risk in control.And the incentive mechanism can well inhibit the malicious behaviors,share the trust information and ensure sensitive information secure during data interaction.

关 键 词：社会网络 信任关系 多维属性 激励机制 

分 类 号：TP31[自动化与计算机技术—计算机软件与理论]