带有可信度的偏序偏好决策信息的聚合方法

An Approach to Aggregating Partially Ordered Preferences with Belief Degrees for Decision Making

提出了一种解决带有可信度的偏序形式的偏好信息的决策方法.对于决策者给出的带有可信度的偏好信息,通过把每个偏好看成一个序列,对于每对方案,利用上下文概率计算一个方案排到另一个方案之前的概率.然后,根据偏好排序算法对这些概率进行聚合,得到一个近似最优的总方案排序.并通过实例说明其有效性.

we propose in this paper an approach to aggregating partially ordered preferences with belief degrees for decision making. We consider a set of preferences associated with belief degrees, and treat each preference as a sequence with belief degree. For every pair of considered alternatives, we calculate the probability that one alternative should be placed before another. These probabilities are then fed to the ORDER-BY-PREFE- ENCE algorithm to generate an approximately optimal total order over all the alternatives. A simple example is given to illustrate this approach.