基于粗集理论的知识系统证据推理研究

证据推理是处理不确定问题的重要方法，但灾用中存在许多问题，如假设的基本概率指派（bpa)往往由专家事先确定，带有较强的主观性，基于证据推理和粗集理论的基本关系，利用粗集约简，决策表确定基本概率指派等方法解决上述问题，并在此基础上，提出了一种决策表的证据推理方法，用于决策表的预测，实例表明，证据推理和粗集理论的结合可提高并对不确定问题的求解能力。

Evidential Reasoning of Knowledge Base System Based on Rough Set Theory

The theory of evidence is important for solving uncertain problems, but there exist some problems . For example, basic probability assignments of evidence are often determined beforehand by experts according to their experience, which seems to be too subjective. To deal with the difficulty, the relationships between rough sets and evidential reasoning are discussed in this paper. Based on the study above, an algorithm for evidential reasoning in decision tables is proposed. An example is also given to illustrate how evidential reasoning integrated with rough sets works. The results show that the integration of rough sets with evidential reasoning is more suitable for uncertain problems.