AKAS—an automated diagnostic knowledge-acquisition system

Abstract In this paper the design and implementation of an automated diagnostic knowledge acquisition system AKAS is described. The system is designed to elicit front-end knowledge from domain experts. AKAS provides a mechanism for interactively interviewing domain experts and transforming results into production rules. The approach taken to develop AKAS was based on an understanding of the causal structure of diagnostic knowledge, and rules are formatted into a knowledge base for use by commercial expert system shells such as LEVEL 5 and INSIGHT 2+. The paper describes the underlying algorithm, the diagnostic knowledge structure, the rule generator and the user interface of AKAS. A detailed example is also given to demonstrate how AKAS works. Several field trials have indicated that the level of interaction provided by AKAS is sufficiently friendly for obtaining up to four levels of causal diagnostic knowledge.

[1]  G. Kelly.  The Psychology of Personal Constructs , 2020 .

[2]  Bruce G. Buchanan,et al.  Artificial Intelligence as an Experimental Science , 1988 .

[3]  T. S. Ng,et al.  A method of knowledge acquisition developed in the construction of DESPLATE , 1988 .

[4]  John H. Boose,et al.  Personal Construct Theory and the Transfer of Human Expertise , 1984, AAAI.

[5]  Larry J. Eshelman,et al.  MOLE: A Knowledge Acquisition Tool that Uses its Head , 1986, AAAI.

[6]  Lawrence M. Fagan,et al.  Use of a Domain Model to Drive an Interactive Knowledge-Editing Tool , 1987, Int. J. Man Mach. Stud..

[7]  Gary S. Kahn,et al.  Strategies for Knowledge Acquisition , 1985, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[8]  Sandra Marcus,et al.  Taking Backtracking with a Grain of SALT , 1987, Int. J. Man Mach. Stud..

[9]  Gary S. Kahn,et al.  A Mixed-Initiative Workbench for Knowledge Acquisition , 1987, Int. J. Man Mach. Stud..

[10]  John H Boose.  ETS—a system for the transfer of human expertise , 1986 .

[11]  Bruce G. Buchanan,et al.  The MYCIN Experiments of the Stanford Heuristic Programming Project , 1985 .

[12]  John H. Boose,et al.  Rapid Acquisition and Combination of Knowledge from Multiple Experts in the Same Domain , 1985, Conference on Artificial Intelligence Applications.

[13]  Antonio Leal,et al.  An Interactive Program Fo Conversational Elicitation of Decision Structures , 1977, IEEE Transactions on Systems, Man, and Cybernetics.