论文部分内容阅读
传统的不确定性推理模型中可信度参数的初始值均由领域内的一个专家给出,由于专家的知识、经验、背景等的不同,其结果必然存在差异,并且随着推理链的延伸,最终可能会得到与事实相悖的结论。为了提高推理的可靠性,提出了通过专家群来确定可信度参数,并给出了评价模型。以疾病诊断专家系统MYCIN知识库中的感冒判断的规则为例,详细介绍了该方法的实际应用过程,证实该方法对提高不确定性推理的可信程度十分必要。
The initial values of the credibility parameters in the traditional uncertainty reasoning model are all given by an expert in the field. Due to the differences in knowledge, experience and background of the experts, the results inevitably have differences, and as the extension of the reasoning chain , May eventually be contrary to the facts and conclusions. In order to improve the reliability of reasoning, we propose to determine the credibility parameters through the expert group, and give the evaluation model. Taking the rule of cold judgment in MYCIN knowledge base of disease diagnosis expert system as an example, the practical application process of this method is introduced in detail, and the method is proved to be very necessary to improve the credibility of uncertainty reasoning.