论文部分内容阅读
不确定性推理问题是专家系统需要解决的基本问题之一,已有的不确定性推理方法在规则的应用要求、加权推理及规则参数获取方法方面具有局限性。该文在分析、归纳、总结不确定性推理相关问题的基础上,提出了基于充分性参数和必要性参数的不确定性推理方法。该方法能够表示具有复杂不确定性加权关系的诊断规则,并且能够根据规则的约束条件,通过数值计算获取规则的不确定性相关参数。该方法在已开发的诊断型专家系统中得到了应用,经验表明该方法是行之有效的。
Uncertainty reasoning is one of the basic problems to be solved by the expert system. The existing Uncertainty reasoning methods have limitations in the application of rules, the weighted inference and the acquisition of rules parameters. On the basis of analyzing, summarizing and summarizing the related problems of uncertainty reasoning, this paper proposes an uncertainty reasoning method based on sufficient and necessary parameters. The method can represent the diagnostic rules with the weighted uncertainty of the complex uncertainties, and can obtain the uncertainty related parameters of the rules by numerical calculation according to the constraint conditions of the rules. The method has been applied in the developed diagnostic expert system, and experience shows that the method is effective.