模块化本体的Tableau算法及其性能优化

Tableau Algorithm of Modular Ontology and Its Performance Optimization

针对现有模块化本体推理方法通用性低、控制复杂等不足,提出一种基于服务的分布式Tableau算法。模块在进行一致性推理时,对关于外部概念的断言,将调用相应模块的服务进行推理,同一推理中的矛盾在定义相应概念的模块中得到捕获,采用优化技术改进算法的时间性能。实验结果表明,该算法使得模块在表述知识时能灵活引用外部概念,支持复杂的推理任务,具有较好的可伸缩性。

The existing reasoning methods of modular ontology are restricted and difficult to manipulate. Aiming at this problem, a distributed service-based Tableau algorithm is presented. During the consistency reasoning, a module can call the services of others to reason asserts about foreign concepts. Every contradiction of a reasoning process is captured in the module defining the related concept. Optimization methods are used to improve the time-efficiency of the algorithm. Experimental results show that by adopting the proposed algorithm, a module can use foreign concepts flexibly to represent knowledge, and some complex reasoning tasks can be fulfilled. The algorithm has good scalability.