大数据量推理专家系统外壳的研究与设计

Study and Design Expert System Shell to Reason on Large Amount of Data

CLIPS是现今主流的专家系统开发工具,但它不能直接访问外部数据库,而传统数据库又缺乏知识。本文提出将CLIPS与数据库系统Postgres结合,使用知识开发技术来降低存储在CLIPS主存储器内数据量。在此结构中,数据存储于Postgres,两系统的接口负责知识归纳,继而将知识传送到CLIPS的工作存储器进行推理。增加了CLIPS的大数据量推理能力。

CLIPS is the popular expert shell. But Clips can not access external dadabase.and traditional dadabas axe short of knowledge. This paper proposes an architecture which integrates CLIPS and Postgres and employing knowledge discovering techniques to reduce the amount of data to be stored in the CLIPS main memory. In this architecture Postgres is coupled with CUPS. The data axe stored into the Postgres. An interface between the two systems is responsible for inducing knowledge from the set of relations. Such induced knowledge is then transferred to the CLIPS working memory. The reasoning ability of CLIPS is advanced in this architecture.