基于不精确推理模型的诊断型专家系统研究

Study of Diagnostic Expert System based on Uncertain Inference Model

介绍了采用VB6 0开发的一个营养诊断专家系统 ,它属于一种诊断型专家系统。该系统的知识表示使用了产生式规则形式 ,推理机采用了规则推理和不精确推理相结合的方法。在不确定性知识处理上 ,应用确定性理论、主观Bayes方法和证据理论分别建立了不精确推理模型 ,使推理机具有较高的准确性。

In this paper we introduce a nutrition diagnostic expert system developed with VB6 0,which is one of the diagnostic expert system.In the system,the method of knowledge representation is genetic rule,inference engine combines uncertain inference and rule inference. In dealing with the uncertain knowledge,we set up the uncertain inference model based on confirmation theory,subjective Bayes method and evidence theory, to raise the accuracy of inference.