基于Rough Sets理论的证据获取与合成方法

Research on rough sets-based evidence acquirement and combination of DST

证据理论是处理不确定性问题的有力工具,它处理的证据来源于专家.专家的知识经验是有限的,获取较困难,且可能存在一定的主观性.针对上述问题,提出了一种基于粗糙集理论的证据获取的新方法,并对证据合成和应用进行了研究.首先研究了大型决策表分解问题.利用粗糙集理论分析条件属性间的依赖关系,对条件属性集进行聚类,形成多个条件属性集相对独立的子决策表;其次对各子决策表进行分析,利用粗糙集的分类思想和隶属度概念,计算证据的基本可信度分配;最后文章对证据的合成及其在决策分析中的应用进行了研究,提出了相应的解决方法.

As a powerful tool in dealing with uncertainty questions, the evidence using in the evidence theory is given by experts. The expert＇s knowledge is limited, subjective and sometimes difficult to obtain. To solve these problems, this paper proposes a new method of knowledge acquirement and presents a method of evidence combination and application. Firstly, we research the problem of how to partition the huge decision table. By analysing of the dependency degree among the attributes based on the Rough Sets （RS）, condition attributes axe classified and the original decision table turns into several small tables, which are independent from one another. Secondly, with the classification method of RS, we analyses the small table, and get the basic probability assignment for evidence theory. Lastly, the problem of how to combine and apply the evidence is discussed and a solution is given as well.