知识网格环境下基于本体的CBR诊断系统架构研究

STUDY ON CBR DIAGNOSIS SYSTEM FRAMEWORK BASED ON ONTOLOGY IN THE ENVIRONMENT OF KNOWLEDGE GRID

提出知识网格环境下基于本体的分布式多案例推理系统架构。该架构引入基于本体的案例推理服务器和任务-方法本体,实现了过程性知识和陈述性知识分离,有效地解决了案例推理重用问题;在案例推理服务器中共享本体约束指导下,各案例库可以灵活地表达各自领域知识,实现知识集成,解决了传统案例推理系统知识局限性问题,实现了各企事业单位之间知识共享和重用。该架构是开放的并采用Web Service技术构建,能够实现网格异构环境下系统集成,具有很强的扩展性。

This paper presents a distributed multiple case-based reasoning（CBR） system framework based on ontology, in knowledge grid environment, which imports case-based reasoning server based on ontology and task-method ontology, realizes the separation of the process knowledge and the statement knowledge, and solves the reuse issue of case-based reasoning effectively. Because domain knowledge in various case bases can be represented flexibly under the constraint and guidance of the share ontology in case-based reasoning server to achieve knowledge integration, the traditional case-based reasoning system knowledge limitation problem can be solved effectively, and the knowledge-share and reuse between departments can be realized. This framework is constructed by Web Service technology and is an open system,it enables the system integration in the grid heterogeneous environment with strong sealability.