论文部分内容阅读
本文提出了一种称为“分类产生式规则”的专家系统知识表示方法，该方法实质性地改进了传统的产生式规则，可有效地解决基于规则的专家系统存在的知识组合爆炸和搜索匹配费用高的问题，同时，较好地解决了专家系统对数值计算无能为力的问题。文中主要介绍了这种方法的定义、表示形式、推理控制策略和知识组织等问题，并简要分析了该方法的优势所在。采用分类产生式规则，已成功地建造了两个实用专家系统。
In this paper, we propose a knowledge representation method of expert system called “classification production rule”, which improves the traditional production rules substantially and effectively solves the knowledge explosion and search matching existing in rule-based expert systems The problem of high cost, at the same time, a better solution to the problem of the expert system can not do numerical calculation. The paper mainly introduces the definition, representation, reasoning control strategy and knowledge organization of this method, and briefly analyzes the advantages of this method. With the rule of classification generation, two practical expert systems have been successfully built.