Skip to main content

The utility problem analysed

A case-based reasoning perspective

  • Conference paper
  • First Online:

Part of the book series: Lecture Notes in Computer Science ((LNAI,volume 1168))

Abstract

In case-based reasoning (CBR) there are compelling arguments in support of large case-bases; greater target problem coverage, better solution quality, improved system efficiency. However a problem known as the utility problem dictates that the last of these arguments is not necessarily true. In fact, adding more cases to an already “saturated” case-base will reduce rather than improve system efficiency. Thus, there is a trade-off situation in which adding cases to improve coverage and quality is pitted against efficiency degradation. This paper discusses the utility problem from a case-based reasoning perspective, examining its root causes in an experimental CBR system.

This is a preview of subscription content, log in via an institution.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

Bibliography

  1. Markovitch, S. & Scott, P. D., Information Filtering. Selection mechanisms in Learning Systems. Machine Learning, 10, 113–151, (1993).

    Google Scholar 

  2. Minton, S., Qualitative Results Concerning the Utility of Explanation-Based Learning. Artificial Intelligence, 42, 363–391, (1990).

    Google Scholar 

  3. Tambe, N., Newell, A., and Rosenbloom, P. S., The Problem of Expensive Chunks and is Solution by Restricting Expressiveness. Machine Learning, 5, 299–349, (1990).

    Google Scholar 

  4. Veloso, M., Learning by Analogical Reasoning in General Problem Solving. Ph.D Thesis (CMU-CS-92-174). Carnegie Mellon University, Pittsburgh, USA, (1992).

    Google Scholar 

  5. Liu, B., Choo, S. H., Lok, S. L., Leong, S. M., Lee, S. C., Poon, F. P., and Tan, H. H., Finding the Shortest Route Using Cases, Knowledge, and Dijkstra's Algorithm. IEEE Expert, 9(5), 7–11, (1994).

    Google Scholar 

  6. Chen, W. K., Theory of Nets: Flows in Networks. John Wiley & Sons, New York. (1990)

    Google Scholar 

  7. Smyth, B. & Keane, M. T., Remembering to Forget: A Competence Preserving deletion Policy for CBR Systems. Proceedings of the 14th International Joint Conference on Artificial Intelligence. Montréal Canada., Morgan Kaufmann, (1995).

    Google Scholar 

Download references

Author information

Authors and Affiliations

Authors

Editor information

Ian Smith Boi Faltings

Rights and permissions

Reprints and permissions

Copyright information

© 1996 Springer-Verlag Berlin Heidelberg

About this paper

Cite this paper

Smyth, B., Cunningham, P. (1996). The utility problem analysed. In: Smith, I., Faltings, B. (eds) Advances in Case-Based Reasoning. EWCBR 1996. Lecture Notes in Computer Science, vol 1168. Springer, Berlin, Heidelberg. https://doi.org/10.1007/BFb0020625

Download citation

  • DOI: https://doi.org/10.1007/BFb0020625

  • Published:

  • Publisher Name: Springer, Berlin, Heidelberg

  • Print ISBN: 978-3-540-61955-0

  • Online ISBN: 978-3-540-49568-0

  • eBook Packages: Springer Book Archive

Publish with us

Policies and ethics