摘要
网络的迅速发展为人们提供了自由、开放的交互方式,互联网服务已经成为大众日常生活的重要组成部分,如何评价表征节点交互的可信度已成为网络应用的核心问题之一。由于开放式网络环境具有匿名性、随机性和动态性等特点,用户在网络中选择目标节点进行交互时面临诸多风险,因此基于节点交互行为的信任评价机制成为抑制网络恶意虚假行为的有效策略机制。首先,概述信任的相关性质,根据网络节点间的交互行为及其产生的信任关系特征,形式化定义与构建信任网络。其次,分析信任机制的框架体系,依次讨论信任机制在P2P网络、电子商务、社会网络中的研究要点与安全威胁。最后,重点对比分析不同领域中的典型信任模型,详述其抗攻击效用与不足,从改进模型算法、提高模型抗攻击能力等方面指出信任机制在未来的研究方向。
The rapid development of Internet provides people with a free and open way of interaction.Internet services have become an important part of populace’s daily life.How to evaluate the trustworthiness of characterizing node interaction has become one of the core issues in network applications.Due to the anonymity,randomness and dynamic characteristics of open network environment,users face many risks in choosing the target node in the network for interaction.The trust evaluation mechanism based on node interaction becomes an effective strategy mechanism to suppress the malicious and fake behaviors of network.First of all,this paper outlined the related properties of trust,and formally defined and constructed the trust network according to the interactive behavior among network nodes and the characte-ristics of their trust relationship.Then,this paper analyzed the framework of trust mechanism,and discussed the research focuses and security threats of trust mechanism in P2P networks,e-commerce and social networks.At last,this paper focused on the comparison and analysis of typical trust models in different fields,detailed the effectiveness and shortcomings of anti-attack,and proposed the research direction of future trust mechanisms in terms of improving the model algorithm and enhancing the anti-attack capability of model.
作者
刘建生
游真旭
乐光学
王结太
刘建华
LIU Jian-sheng;YOU Zhen-xu;YUE Guang-xue;WANG Jie-tai;LIU Jian-hua(Department of Science,Jiangxi University of Science and Technology,Ganzhou,Jiangxi 341000,China;Department of Mathematics and Information Engineering,Jiaxing University,Jiaxing,Zhejiang 314000,China)
出处
《计算机科学》
CSCD
北大核心
2018年第11期13-28,36,共17页
Computer Science
基金
国家自然科学基金项目(61462036
61572014
61702224)
浙江省自然科学基金项目(LY16F020028
LQ15F010008
LY15F020040)资助
关键词
网络交互
恶意攻击
信任机制
信任模型
Network interaction
Malicious attack
Trust mechanism
Trust model
作者简介
刘建生男,副教授,CCF会员,主要研究方向为深度学习;游真旭女,硕士生,CCF学生会员,主要研究方向为复杂网络理论;乐光学男,教授,CCF会员,主要研究方向为复杂网络、协同服务等,E-mail:guangxueyue@163.com(通信作者);王结太男,博士,主要研究方向为无线网络技术;刘建华男,副教授,主要研究方向为网络安全。
