Knowledge Acquisition for Expert Systems

Building an expert system involves eliciting, analyzing, and interpreting the knowledge that a human expert uses when solving problems. Expe­ rience has shown that this process of "knowledge acquisition" is both difficult and time consuming and is often a major bottleneck in the production of expert systems. Unfortunately, an adequate theoretical basis for knowledge acquisition has not yet been established. This re­ quires a classification of knowledge domains and problem-solving tasks and an improved understanding of the relationship between knowledge structures in human and machine. In the meantime, expert system builders need access to information about the techniques currently being employed and their effectiveness in different applications. The aim of this book, therefore, is to draw on the experience of AI scientists, cognitive psychologists, and knowledge engineers in discussing particular acquisition techniques and providing practical advice on their application. Each chapter provides a detailed description of a particular technique or methodology applied within a selected task domain. The relative strengths and weaknesses of the tech­ nique are summarized at the end of each chapter with some suggested guidelines for its use. We hope that this book will not only serve as a practical handbook for expert system builders, but also be of interest to AI and cognitive scientists who are seeking to develop a theory of knowledge acquisition for expert systems.

1 Knowledge Acquisition—An Introductory Framework.- 1. Introduction.- 2. The Knowledge Acquisition Process: Mining Is a Misguided Analogy.- 3. Knowledge Domains: Capturing or Creating a Language.- 4. Tasks: Defining the Problem That the Expert System is Designed to Solve.- 5. System Modality: Defining the Set of Tasks.- 6. Users: Acquiring Their Knowledge.- 7. An Alternative Approach: Identifying Weaknesses in Human Reasoning.- 8. Concluding Guidelines for Knowledge Engineers.- 9. Brief Overview of Chapters.- 10. References.- 2 Use of Models in the Interpretation of Verbal Data.- 1. Problems and Solutions in Building Expert Systems.- 2. A Methodology for Knowledge Acquisition.- 3. Interpretation of Data on Expertise.- 3.1. Domains and Tasks.- 3.2. Levels of Interpretation.- 3.3. Primitives of the Epistemological Level.- 3.4. Interpretation Models.- 3.5. Use of Interpretation Models.- 3.6. Supporting Knowledge Acquisition: KADS and ROGET.- 4. Concluding Remarks.- 5. Guidelines Summary.- 6. References.- 3 Knowledge Acquisition by Analysis of Verbatim Protocols.- 1. Introduction.- 2. Design of the Experiment.- 3. The Nephrotic Syndrome.- 4. Analysis of the Transcript.- 5. The Domain Model—Structural Description.- 6. Qualitative Simulation in the Explanation.- 7. The Domain Model—Qualitative Description of State.- 8. The Domain Model—Qualitative Simulation.- 9. Conclusion.- 10. Postscript.- 10.1. Computer Implementation.- 10.2. Types of Analysis.- 10.3. Guidelines Summary.- 11. References.- 4 A Systematic Study of Knowledge Base Refinement in the Diagnosis of Leukemia.- 1. Introduction.- 2. Leukemia Diagnosis.- 3. Knowledge Elicitation and Expert System Development.- 3.1. Expert System Package.- 3.2. System Development.- 3.3. Knowledge Base Development.- 4. System Performance.- 5. Analysis of Disagreements between System and Expert.- 6. Discussion.- 7. Guidelines Summary.- 7.1. Limits of the Methods Proposed.- 8. References.- 5 Knowledge Elicitation Involving Teachback Interviewing.- 1. The Knowledge Elicitation Process.- 1.1. Theoretical Stance.- 1.2. Conversation Theory.- 2. Case Studies.- 2.1. Why Use Teachback Interviewing?.- 2.2. Arithmetic Study.- 2.3. VLSI Design Study.- 3. Mediating Representation—SGN.- 4. Discussion.- 4.1. Teachback as a Complete Methodology.- 4.2. Teachback Interviewing as a Viable Technique.- 5. Guidelines Summary.- 5.1. Strengths.- 5.2. Weaknesses.- 5.3. Rules of Thumb.- 6. References.- 6 An Interactive Knowledge-Elicitation Technique using Personal Construct Technology.- 1. Knowledge Engineering.- 2. Personal Construct Psychology.- 3. What Is a Repertory Grid?.- 3.1. Eliciting Constructs.- 4. Techniques for Repertory Grid Elicitation and Analysis.- 4.1. Repertory Grid Analysis.- 4.2. Analysis of a Single Grid.- 4.3. Analysis of a Pair of Grids.- 4.4. Analysis of a Group of Grids.- 5. Soft Systems Analysis.- 5.1. The Significance of Different Perspectives.- 5.2. Techniques of Soft Systems Analysis.- 6. PLANET: A Computer-Based Knowledge-Engineering System.- 7. PEGASUS in Action.- 8. ENTAIL in Action.- 9. Validation.- 10. Guidelines Summary.- 11. References.- 7 Different Techniques and Different Aspects on Declarative Knowledge.- 1. Introduction.- 2. Methods.- 2.1. Concept Elicitation.- 2.2. Structure Elicitation.- 2.3. Structure Representation.- 2.4. Developing the Representation.- 3. Using the Knowledge Base.- 4. Future Research.- 5. Guidelines Summary.- 6. References.- 8 Role of Induction in Knowledge Elicitation.- 1. Introduction.- 2. Induction.- 2.1. General Principles.- 2.2. The ID3 Algorithm.- 3. A Case Study.- 3.1. Background and Rationale.- 3.2. The Knowledge Domain.- 3.3. Procedures.- 3.4. Summary of Findings.- 3.5. Interviewing the Expert.- 3.6. Comments on the Interviews.- 4. Conclusion.- 4.1. Issues in Induction.- 4.2. Guidelines Summary.- 5. References.