基于本体的分布式实例推理技术研究

为了克服单一实例库知识的局限性,实现分布式环境下多数据源的知识重用和共享,提出了一个分布式实例推理系统框架.系统通过本体服务器建立和维护实例库之间的本体知识,其中基本本体为知识的表示提供了全局约束和基础,实例推理服务器可以在基本本体框架下定义领域本体来灵活表达各自的领域知识,而本体目录则为知识的检索提供了向导.本体的引入解决了不同实例库之间知识的互理解和互操作性,能够有效地实现多实例库的协同推理.系统采用Web Service技术构建,是一个开放的系统框架,具有很强的可扩展性.

Research on Ontology-based Distributed CBR Technology

To cope with the limitation of single case base knowledge, and realize knowledge reuse and sharing of multiple data source in distributed environment, a distributed cased-based reasoning system framework is proposed in this paper. In this system, ontology knowledge of the case bases is constructed and maintained by ontology server, base ontology provides global constraints and building blocks for knowledge representation, CBR server can define domain ontology to represent its domain knowledge under the base ontology framework, moreover, ontological directory provides guidance of knowledge retrieval. The introduction of ontology solves the inter-understanding and inter-operation of different case base, and therefore it realizes multi-case-based reasoning collaboratively. The system is developed using web service technology, and it is an open and flexible framework.