CBR in a changing environment

  • D. Y. Joh
Application Papers
Part of the Lecture Notes in Computer Science book series (LNCS, volume 1266)


Case-Based Reasoning (CBR) has been proposed for design tasks in which past experience is exploited to solve the current problem. Based on a study of experts, it is believed that a case based approach would be appropriate as the basis for computer aided decision support system for internetwork design. However, certain characteristics of the internetwork design domain require that the state of the art in CBR be extended before it could be applied to internetwork design.

A knowledge revision mechanism is proposed to extend the use of previous cases. Knowledge revision updates information about design components and uses that information to augment the case base, enabling the retrieval mechanism to select both from actual experiences and from experiences which might have occurred had current devices been available at the time. A computer program, CIDA, implements key portions of the model.

An empirical experiment was performed to validate the model. The results, blinded, were graded by three evaluators. A statistical analysis of the evaluations indicates that CIDAs performance is between that of experts and intermediates, but is significantly better than that of human novices. An ablation experiment shows the extended CBR approach has advantages over both existing state-of-the-art CBR systems and constraint satisfaction systems.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. Cohen, P. and Howe, A. (1988). “How Evaluation Guides Research, “AI Magazine, 9(4), 35–43.Google Scholar
  2. Ericsson K. A. and Simon, H. A. (1984). Protocol Analysis: Verbal Reports as Data, Cambridge, MA, MIT Press.Google Scholar
  3. Joh, D. (1996). “Knowledge Revision in Case-Based Reasoning: A Cognitive Model of the Telecommunications Internetwork Design Process,” Ph.D. Dissertation, Joseph M. Katz Graduate School of Business, University of Pittsburgh.Google Scholar
  4. Kolodner, J. (1991). “Improving Human Decision Making through Case-Based Decision Aiding,” AI Magazine, 12(2), 52–68.Google Scholar
  5. Schank, R. C. (1982). Dynamic Memory, Cambridge University Press, Cambridge, UK.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1997

Authors and Affiliations

  • D. Y. Joh
    • 1
  1. 1.Handong UniversityPohang, KyoungbukKorea

Personalised recommendations