一种知识表达Petri网及其特性分析

A Knowledge Representing Petri Net and Its Properties

根据Horn子句的特点,定义了一种新的Petri网形式———HIPN网,给出了基于HIPN网的知识表达方式和从逻辑程序到HIPN模型的转换算法,论证了HIPN网的一些重要特性.基于HIPN模型的逻辑推理过程和现有求解方法相比,可以大幅减少重复推理,有效提高推理效率.

There have been a lot of studies in the field of knowledge represented by Petri net, but there still exist many problems to be solved. Finding the best way to represent knowledge, maintaining the knowledge base correctly and reasoning in fast speed are all problems that should be studied carefully. According to the properties of Horn sentence, a new kind of Petri net called HIPN net is defined. Some important properties are shown and proved. How to represent knowledge by HIPN and invent an algorithm to transform a logical program into a HIPN model are discussed. The results show that HIPN is a suitable model for knowledge representing with great advantages. Redundant calculations have been reduced and speed has been increased by reasoning in HIPN.