Building expert systems
Introductory Readings in Expert Systems
Introductory Readings in Expert Systems
| Hi-index | 0.00 |
As more (and more) knowledge-based systems are designed for operational use, it becomes apparent that the knowledge engineering (KE) process of transferring domain expertise into the system consumes the single largest block of development time (Feigenbaum, 1977; Hayes-Roth, Waterman, & Lenat, 1983). To reduce the time required for this stage, numerous tools have been developed that automate or systematize the elicitation and transfer of knowledge from expert to knowledge base. Among these advances are developments in machine learning that enable an expert system to acquire its knowledge from "experience" (Michalski, 1980; Michie, 1982), expert system shells that allow domain experts to write their own rules (e.g., MacSMARTS), and knowledge engineering tools that assist a knowledge engineer in his/her work. The material reported here describes BDM's development of a computer-based set of tools, known collectively as BDM-KAT.