Brought to you by EarthWeb
IT Library Logo

Click Here!
Click Here!

Search the site:
 
EXPERT SEARCH -----
Programming Languages
Databases
Security
Web Services
Network Services
Middleware
Components
Operating Systems
User Interfaces
Groupware & Collaboration
Content Management
Productivity Applications
Hardware
Fun & Games

EarthWeb Direct EarthWeb Direct Fatbrain Auctions Support Source Answers

EarthWeb sites
Crossnodes
Datamation
Developer.com
DICE
EarthWeb.com
EarthWeb Direct
ERP Hub
Gamelan
GoCertify.com
HTMLGoodies
Intranet Journal
IT Knowledge
IT Library
JavaGoodies
JARS
JavaScripts.com
open source IT
RoadCoders
Y2K Info

Previous Table of Contents Next


Part III
Critical Technologies Associated with Expert Systems

Chapter 11
Case-Based Reasoning

Lean Suan Ong and Arcot Desai Narasimhalu


CONTENTS

1. Introduction
2. History of CBR
3. CBR Concepts
3.1. Representing a Case
  3.2. Indexing Cases
  3.3. Similarity Matching
  3.4. Adaptation
4. CBR Applications
4.1. SMART System
  4.2. The CARES System
5. CBR Tools
6. Research Challenges
6.1. Case Representation
  6.2. Case Indexing
  6.3. Case Adaptation and Learning
7. Emerging Hybrid Systems
7.1. CBR and Rule-Based Reasoning (RBR)
  7.2. CBR and Information Retrieval
  7.3. CBR in the Case of Multiple Media
  7.4. CBR and Model-Based Reasoning
8. Conclusion
References

1. INTRODUCTION

No matter what type of problem-solving task it is, people usually have more confidence in you when you say that you have done that particular task many times before, i.e., you have the experience. Why? Because by having done that task before, it means that you are able to do another similar task, better and faster, with less potential for failure; you are an "expert" in that problem-solving task. It is also a well-known fact that human experts solve problems by relying on their past experiences in solving similar problems. This is especially true in areas such as law and medicine, for example. Lawyers rely on previous cases to argue for or against a new case, just as doctors rely on previous patient cases that they have seen for diagnosis and treatment of a new patient. It was from this idea of reusing past experiences to solve new or current problems that an Artificial Intelligence (AI) approach called Case-Based Reasoning (CBR) was born. CBR is sometimes classified under Machine Learning, and supports knowledge acquisition and problem-solving. It is also sometimes associated with other technologies such as analogy, cognitive psychology modeling, machine learning, and information retrieval, for example.

2. HISTORY OF CBR

CBR by itself is not new technology; it has its origins in the work of Roger Shank on Dynamic Memory (Schank, 1982), which described a memory-based approach to reasoning. His ideas were further expanded by graduate students at Yale University and Janet Kolodner of Georgia Tech, who developed the CYRUS System (Kolodner, 1983). A series of annual workshops have been held since 1987, and it is workshops such as the CBR workshop sponsored by DARPA (Defense Advanced Research Project Agency) at Florida in 1989 that brought together the CBR researchers and spurred interest in this technology. Interest and participation in CBR has grown over the recent years, and there are conferences and workshops held regularly (such as the European Workshop on CBR (EWCBR)) that are very well received.

Early CBR systems developed include CHEF (Hammond, 1986) a meal planning system; PROTOS (Bareiss, 1988) for classifying and thus diagnosing hearing disorders; MEDIATOR (Simpson, 1985) in the domain of dispute mediation; and HYPO (Ashley and Rissland, 1988), which provides arguments for and against parties in U.S. nondisclosure litigation, behaving very much like a lawyer would. However, it is the availability of commercial CBR tools and success of help-desk applications such as the SMART (Acorn et al., 1992) system for Compaq computers that have given the impetus to this technology.


Previous Table of Contents Next

footer nav
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.