计算机应用 ›› 2012, Vol. 32 ›› Issue (08): 2350-2359.DOI: 10.3724/SP.J.1087.2012.02350

• 典型应用 • 上一篇    下一篇

选择在线交易服务者的多情境信任和声誉评估系统

刘彬,张仁津   

  1. 贵州师范大学 数学与计算机科学学院,贵州 贵阳550001
  • 收稿日期:2012-02-20 修回日期:2012-04-11 发布日期:2012-08-28 出版日期:2012-08-01
  • 通讯作者: 刘彬
  • 作者简介:刘彬(1973-),男,湖南华容人,副教授,硕士,CCF会员,主要研究方向:可信计算、电子商务、网络计算;
    张仁津(1963-),男,四川成都人,教授,主要研究方向:软件方法学、组件技术、对象技术。
  • 基金资助:
    贵州省科学技术基金资助项目 (黔科合J字[2010]2102号)

Multi-context trust and reputation evaluation system for server selection in online transaction

LIU Bin,ZHANG Ren-jin   

  1. School of Mathematics and Computer Science, Guizhou Normal University, Guiyang Guizhou 550001, China
  • Received:2012-02-20 Revised:2012-04-11 Online:2012-08-28 Published:2012-08-01
  • Contact: LIU Bin

摘要: 在以信任和声誉值选择服务者的系统中,普遍存在的问题是在评估服务者的信任和声誉值时考虑因素过少,局限于系统内,以及方法缺少灵活性,导致选择的服务者不能满足用户多样性的要求。为了解决这个问题,提出一种应用于在线交易服务者选择的多情境信任和声誉评估系统,服务者通过注册的质量属性和担保资金获得信任和声誉启动值,由系统内、外的交易经历获得信任和声誉经验值,实际信任和声誉值由两者动态线性组合,后者所占比重随交易次数的增加而动态增加,根据用户设置的服务者的情境属性与实际信任和声誉值选择服务者。通过与其他方法对比测试,结果表明此方法更容易满足用户的各种要求。此信任和声誉评估系统不仅能给新服务者与其他服务者一个公平竞争的环境,也可以减少用户选中恶意服务者的机会。

关键词: 电子商务, 声誉, 信任, 信任评估, 在线交易, 服务者

Abstract: In the systems which select servers according to reputation and trust values, the common problems, which result in that the servers chosen cannot meet the users' diversity requirement, are a small number of factors being considered and limited in the system and lack of flexibility in methods. In order to resolve these problems, a multi-context reputation and trust evaluation system, which was used for selecting servers in online transaction, was proposed. A server acquired its reputation and trust bootstrapping value through its registered quality attributes and guarantee fund, gained its reputation and trust experimental value by its trade experience in the internal system and external systems. The actual trust value was the dynamic linear combination of both values. With the increase of transaction times, the weight of the latter increased dynamically. Server was selected according to the context quality attribute set by user and the actual trust and reputation value. This selection method of a new server was tested and contrasted with other trust and reputation systems. The experimental results indicate that this method can easily meet the requirements of all kinds of users. This trust and reputation could afford the new server as well as other servers a competitive environment, and also reduced the possibility that user selected malicious server.

Key words: electronic commerce, reputation, trust, trust evaluation, online transaction, server

中图分类号: