An object knowledge canonical form for knowledge reuse

被引：4

作者：

Huarng, KH ^{[1
]}

Simmons, DB ^{[1
]}

机构：

[1] TEXAS A&M UNIV, DEPT COMP SCI, COLLEGE STN, TX 77843 USA

来源：

EXPERT SYSTEMS WITH APPLICATIONS | 1996年 / 10卷 / 01期

关键词：

D O I：

10.1016/0957-4174(95)00040-2

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

The importance of knowledge reuse and sharing has been widely recognized. An Object Knowledge Canonical Form (OKCF) is created to translate object knowledge among commercial object-oriented expert system building tools (ESBTs): CLIPS 5.0, KAPPA, and KEE. The OKCF is represented in Backus-Naur Form (BNF), including identifiers, data types, data facts, simple facts, tuple facts, classes of facts, and structured facts. Following these BNF rules, translators have been implemented. These translators can be used to demonstrate the feasibility of OKCF. The use of OKCF in knowledge translation is effective. Besides, OKCF can serve as a reference for the development of new ESBTs. The OKCF can be integrated with knowledge acquisition tools for expert system prototyping. The OKCF is compared with Knowledge Interchange Format in various aspects: commitment to commercial ESBTs, support of important features, and applications.

引用

页码：135 / 146

页数：12