基于移动Agent的智能交易安全风险控制机制研究  

Research on Intelligent Transaction Risk Control Mechanism Based on Mobile Agent System

作　　者：蒋伟进 吕斯健 张莲梅[2] 张婉清 李恩 杨莹 JIANG Wei-jin;LV Si-jian;ZHANG Lian-mei;ZHANG Wan-qing;LI En;YANG Ying(School of Computer Science,Hunan University of Technology and Business,Changsha 410205,China;Electrical Engineering College,Wuhan University,Wuhan 430072,China;College of Frontier Intersection,Hunan University of Technology and Business,Changsha 410205,China)

机构地区：[1]湖南工商大学计算机学院,湖南长沙410205 [2]武汉大学电气工程学院,湖北武汉430072 [3]湖南工商大学前沿交叉学院,湖南长沙410205

出　　处：《模糊系统与数学》2022年第1期144-156,共13页Fuzzy Systems and Mathematics

基　　金：国家自然科学基金资助项目(61772196;61472136);湖南省自然科学基金重点项目(2020JJ4249);湖南省社会科学基金重点项目(2016ZDB006);湖南省社会科学成果评审委员会课题(湘社评19ZD1005);湖南省学位与研究生教育改革研究基金资助项目(2020JGYB234)。

摘　　要：在移动网络交易中,交易双方的互信是网络交易能够顺利进行的前提和关键,实体间的信任度量涉及到交易额、交易时间、消费实体个人相关信息及其对信任的风险态度等因素,通常难以准确地进行量化计算。为探明这种互信关系的本质特点,基于多Agent系统协同理论,提出了基于移动Agent的动态互信计算模型,实现了对信任的定性与定量的转换;同时对现有的新用户信任度赋值方法进行了改进,提出一种依据前期系统的最小信任度,动态设置新用户初始信任度的方法,有效抵制了随意抛弃信誉资料的败德行为。并通过将交易评价体系和权重体系引入多因素机制,增强了互信算法在交易单个属性上的敏感性,有效的过滤和清除了在信誉榨取、信誉诽谤等恶意行为方面对交易信任的影响。详细的理论分析和大量仿真实验验证该机制能有效揭示和解决移动网络交易中的信任计算问题,为移动互联网环境下安全交易的研究提供了有价值的新思路。On-line transactions of the mobile commerce, trust is the premise and key, which makes transactions smoothly. The measurement of trust between entities involves in trading volume, transaction time, personal income of consumer entity and its risk attitude to trust, and so on, which is difficult to be given accurately quantitative calculation.To find out the nature of the trust relationship, based on multi-agent system coordination theory, a dynamic trust calculation model based on mobile Agent is given. The model can achieve the qualitative and quantitative conversion of trust. To improve the existing assignment method of new-user-trust-value, we proposed a system on the based on previous minimum trust, dynamic set the method of the new user’s initial trust, effectively resist the arbitrary abandonment of reputation data. By using the multi-factor mechanism in the introduction of evaluation system and weight system, and enhances the sensitivity of the trust algorithm on a single property of transaction, and effectively filter and eliminate the influence on transaction trust in malicious behaviors such as reputation extraction and reputation defamation. The detailed theoretical analysis and simulation results show that the mechanism can effectively reveal and resolve the trust computing problem of mobile network transactions, and provides a valuable new idea to transaction security in MI.

关 键 词：移动互联网 动态信任模型 信誉管理 移动AGENT系统 在线交易安全 

分 类 号：TP273[自动化与计算机技术—检测技术与自动化装置]