![]() |
|||
![]()
|
![]() |
![]() |
![]() |
Chapter 9
|
1. | Introduction | |
2. | Background and History | |
3. | Techniques and Methodologies | |
3.1. | Introduction | |
3.2. | Model Categories | |
3.3. | Qualitative Models | |
3.4. | Qualitative Simulation | |
3.5. | Model-Based Diagnosis | |
4. | Applications of MBR | |
4.1. | Introduction | |
4.2. | Where is MBR Used? | |
4.3. | Monitoring | |
4.4. | Control | |
4.5. | Diagnosis | |
5. | Trends and Open Issues | |
6. | Summary | |
References |
Most early expert systems relied on a knowledge base acquired from domain experts through an elaborate knowledge acquisition process. This was a consequence of the insight that "knowledge is power," applied to artificial intelligence (AI) systems, and was a key factor in the rise and success of expert systems. However, the knowledge acquisition approach led to systems with inadequate behavior in situations not covered by the expert's experience. The knowledge bases developed in this manner were often difficult to maintain, and gradually became obsolete as the target domain inevitably changed. This partly explains the many cases of otherwise successful early expert systems that gradually fell out of use.
As demonstrated in other chapters of this book, many successful responses to the knowledge acquisition challenge have been developed, including improved knowledge acquisition methodologies and tools, machine learning, and so on. In this chapter, we are concerned with another response, the so-called model-based reasoning (MBR) approach. In this approach, the expert system is seen as composed of a model of the target domain, which is used by the MBR engine to solve problems of interest to the expert system's users. The key point is that the model does not just embody an expert's heuristic experience, but somehow is a more "first-principles" description of the target domain.
In subsequent sections of this chapter, we will further motivate the MBR approach by referring to its historical development as part of AI, followed by an overview of techniques and methodologies that are available for MBR systems. We will describe some categories and examples of applications of the MBR technology. Finally, we will offer some ideas on future trends and open issues.
Previous | Table of Contents | Next |
![]() |
|
Use of this site is subject certain Terms & Conditions. Copyright (c) 1996-1999 EarthWeb, Inc.. All rights reserved. Reproduction in whole or in part in any form or medium without express written permission of EarthWeb is prohibited. Please read our privacy policy for details. |