• 中国计算机学会会刊
  • 中国科技核心期刊
  • 中文核心期刊

J4 ›› 2015, Vol. 37 ›› Issue (09): 1768-1776.

• 论文 • 上一篇    下一篇

一种多维集成直觉模糊信息的信任评价方法

徐军1,钟元生1,郑也夫2   

  1. (1.江西财经大学信息管理学院,江西 南昌 330013;2.华东交通大学经济管理学院,江西 南昌 330013)
  • 收稿日期:2014-07-04 修回日期:2014-11-11 出版日期:2015-09-25 发布日期:2015-09-25
  • 基金资助:

    国家自然科学基金资助项目(71361012,61263018);江西省自然科学基金资助项目(20132BAB201050);江西财经大学校级课题资助项目

A trust evaluation approach of multidimensional
integrated intuitionistic fuzzy information  

XU Jun1,ZHONG Yuansheng1,ZHENG Yefu2   

  1. (1.College of Information Management,Jiangxi University of Finance and Economics,Nanchang 330013;2.School of Economics and Management,East China Jiaotong University,Nanchang 330013,China)
  • Received:2014-07-04 Revised:2014-11-11 Online:2015-09-25 Published:2015-09-25

摘要:

信任是保证在线服务质量的前提。针对电子商务环境存在的复杂性、模糊性和不确定性等因素,引入直觉模糊理论,提出了一种多维集成直觉模糊信息的信任评价方法,给出了两类信任反馈属性集成直觉模糊信息的方法,通过直觉模糊熵确定属性权重,并利用直觉加权算数平均算子求出实体的综合直觉模糊数,从而全面、客观和细腻地反映信任的模糊性和不确定性。计算实例和仿真实验表明,基于多维集成直觉模糊信息的信任评估方法是可行的,通过对风险偏好因子的灵敏度分析,发现服务请求者风险偏好不同,则评估的服务提供商的信任得分也不同;同时也验证了该方法可以有效地抑制恶意节点的攻击,为电子商务环境下信任评价的研究提供了新思路。

关键词: 直觉模糊, 信任评价, 反馈属性, 风险偏好, 信任得分

Abstract:

Trust provides guarantee for quality of service. In e-commerce environment,factors such as complexity,fuzziness and uncertainty can’t ensure consumer payment.It is urgent to judge service providers’credibility and ability to provide services. In this paper,taking the ecommerce environment as the research background,by use of intuitionistic fuzzy (IF) theory,we propose a trust evaluation approach of multidimensional feedback IF information.This trust evaluation approach can implement conversions between feedback attributes and IF sets.We obtain the weight of the attributes via entropy for intuitionistic fuzzy sets,and then aggregate the IF information corresponding to each entity through intuitionistic weighted arithmetic averaging operators,which comprehensively,objectively,and delicately reflects the fuzziness and uncertainty of the trust.Examples and simulation results show the feasibility of the trust evaluation approach.A sensitivity analysis for risk appetites shows that different risk appetites of service requestors will lead to different trust scores of service providers under estimation. It is also verified that the proposed approach can effectively inhibit attacks from malicious nodes, and it may provide a new way for research on trust evaluation in ecommence environment.

Key words: intutionistic fuzzy (IF);trust evaluation;feedback attribute;risk appetite;trust score