Abstract: In order to address the problem that most current reputation models cannot build trust chain efficiently，this paper proposes a trust transfer method based on feedback’s departure degree.It considers the average departure degree of the peers giving feedback as their distance’s ruler.To those peers having no common partners，it uses triangle inequality to calculate their global average departure degree through their trust path built by other peers.Then it finishes the trust transfer and can combine the peer’s direct experience and indirect experience together.The simulation results demonstrate that this trust transfer method is effective.

Key words: trust transfer, departure degree, triangle inequality

摘要： 针对目前声望模型普遍无法有效建立信任链的问题，提出了一种利用评价偏离度进行信任传递的方法。该方法利用平均偏离度作为评价者之间距离度量的尺度，对两个没有共同交互伙伴的评价者，以其他评价者为中间节点构造信任路径，并利用三角不等式计算二者之间的全局平均偏离度，从而进行直接经验与间接经验的融合，实现信任传递。仿真结果表明：该信任传递方法是行之有效的。

关键词: 信任传递, 偏离度, 三角不等式