摘要： 在实际可信网络中现有评估方法的计算复杂度较高，且不易通过编程实现。为此，提出一种新的用户行为评估方法，对网络用户建立用户行为证据优属度，利用模糊决策分析理论确定用户行为证据权重，进而得出用户行为可信隶属度。在此基础上，实现用户行为评估。仿真实验结果表明，该方法的计算复杂度较低，易于通过编程实现，能够用于网络用户评估。

关键词: 可信网络, 用户行为, 模糊决策分析, 隶属度

Abstract: Much more user behaviors evidence is needed in user behaviors assessment in trusted network. Low computational complexity and easily programmable are necessary to an assessment approach. Previous methods should be improved in the two performances. A new method is proposed to assess user behaviors. The optimal membership degree of user behaviors evidence and the weight of user behaviors evidence determined by fuzzy decision analysis theory are used to calculate the trust membership degree of the user behaviors. User behaviors are assessed by their trust membership degree. Experimental results show the method can be used to assess the practical network user behaviors availably both in the low computational complexity and easily programmable.

Key words: trusted network, user behavior, fuzzy decision analysis, degree of membership

中图分类号: 

• N945