普适服务中基于模糊神经网络的信任测度方法

以提供高信任度的主动伺候式普适服务为目标，研究基于模糊神经网络的信任测度问题．首先提出一种基于Agent封装的模糊神经网络结构，它能主动发现服务并进行自发互操作，多Agent之间能相互协调和协同工作；然后提出一种基于模糊神经网络的信任测度方法，它能克服传统方法存在的梯度误差大而导致信任有效性较低的缺陷，在占用较少空间的前提下较大地提高了信任度．实验结果表明，所设计的方法是正确而有效的．

Approach of Belief Measure Based on Fuzzy-neural Network for Proactive Service

To achieve high belief degree proactive/attentive pervasive service, the belief measure problem with new character based on fuzzy-neural network is studied. A kind of new belief measure structure, called fuzzy-neural network structure under the encapsulation of agent is presented, which can discover service proactively and supply spontaneous interoperation in cooperation with multi-agents. Based on this structure, a new fuzzy-neural-based method for belief measure is put forward, which can overcome the shortcoming of traditional approaches with lower belief degree under less extra space caused by larger grads error. Scenarios of implemented demo show the correctness and efficiency of the method.