ABSTRACT
The acquisition of knowledge remains the critical phase of expert system development. Illustrative of the first class of support tools is the SALT system which has built-in knowledge of the problem solving strategies used when configuring electro-mechanical devices. Knowledge engineering workbench would implement a range of elicitation techniques, incorporate machine learning methods, be applicable across domains, and embody a principled or knowledgeable approach to the entire acquisition process. Interpretation models are knowledge constructs at the inference layer–they indicate how components of expertise are organised and used in problem solving. Knowledge about the knowledge acquisition tools themselves constitute a third type of knowledge to be used in any knowledge-based system for knowledge acquisition. The directive model contains knowledge which can be used to inform acquisition. In laddering the expert and the knowledge engineer construct a graphical representation of domain knowledge in terms of the relations between knowledge elements.
