领域知识语义距离及其在专家系统中的应用

Domain knowledge semantic distance and its application in expert system

面向专家系统中的可程序化知识间距离的测度,在构造领域知识本体及本体树的基础上,建立知识身份距离、知识概念距离、知识属性距离、知识描述距离四个测度,并在规范化后采用有序加权几何算子对其加以集结,形成一个更加全面且应用性强的领域知识语义距离模型,有效地解决了专家系统中领域知识的判别问题。实验表明:该模型可快速地判别专家系统中两领域知识的相似程度,并具有82%以上的评判正确率。

Oriented to the measurement of programmable knowledge distance in the expert system,based on the domain knowledge ontology and ontology tree,the paper established four measurements： knowledge identity distance,knowledge concept distance,knowledge attribute distance,and knowledge description distance,which were assembled by Ordered Weighted Geometric（OWG） operator after being standardized.As a result,a more comprehensive domain knowledge semantic distance model with strong applicability was obtained,so as to effectively solve the identification problem of domain knowledge database in expert system.The results show that： this model can quickly determine the similarity of two domain knowledge in expert system with an accuracy of above 82%.