摘要
如何安全、有效地选择信任度满足要求的服务提供者是目前信任研究领域的热点问题。文中提出了一种多属性决策的层次化信任模型,将决策属性分为定量属性和定性属性,有利于属性的分类度量、属性的系统化分析和信任等级的设定,同时给出了对具有不同信任等级的属性进行融合的方法。用置信区间来表示定性属性,可以很好地表达主观不确定性和风险意识。利用推荐者的推荐信息可以快速建立实体间的信任关系,针对网络中存在的恶意实体的推荐,列举了检验推荐信息可信度的方法并分析了各自的优缺点。介绍了几种典型的信任模型并提出了建立信任模型的关键。
Currently, it is a hot issue in trust research area of how to select the trust- satisfied service providers safely and effectively. A hierarchical trust model was provided through dividing decision attributes into quantitative attributes and qualitative attributes. This was beneficial to the classified measurements of the attributes, systematic analysis of the attributes and the setting of the decision levels. Then, the method of combining different trust levels was explained. Using confidence interval to indicate qualitative attributes could express subjective uncertainty and risk consciousness better. The trust relationships between entities could be established quickly by using the recommendations from recommenders. For the existing malicious recommendations, inspecting methods were listed accompanying with the respective advantages and disadvantages. At last, several typical trust models were introduced and the key issues of how to establish trust models were proposed.
出处
《计算机技术与发展》
2010年第3期36-39,43,共5页
Computer Technology and Development
关键词
信任模型
属性分类
推荐信任
trust model
attributes classification
recommended trust
