Abstract
Associational reasoning solves common problems quickly. Model-based reasoning can be used to solve unfamiliar, unusual, or difficult problems, but it does so slowly. This paper describes a method for increasing the solution speed of model-based reasoning systems by remembering previous similar problems and making small changes to their solutions. This technique is known as case-based reasoning (CBR). If the CBR system were limited to solving only problems identical to those it had seen before, this technique would be nothing more than caching. However, by integrating a model-based component with CBR, a system can make model-based modifications to previous solutions to fit the details of similar but not identical problems. Only parts of the solution that depend on features which differ in the old and new problem must be modified. Therefore, the computational cost of arriving at the CBR solution is dependent on the magnitude of the difference between the new problem and the retrieved case, and not on the complexity of the problems themselves. For problems that are computationally expensive to solve, then, case-based reasoning has the potential to make a significant improvement in solution speed. The implementation of such a system for the domain of medical diagnosis is presented, and its extension to other domains is described.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Davis, R. Expert systems: where are we? and where do we go from here? AI Magazine, 3:2 (1982) 3–22
Davis, R. Diagnostic reasoning based on structure and behavior. Artificial Intelligence, 24 (1984) 347–410
Kolodner, J. Maintaining organization in a dynamic long-term memory. Cognitive Science, 7 (1983) 243–280
Kolodner, J., Simpson, R., Sycara-Cyranski, K. A process model of case-based reasoning in problem solving. In Proceedings of the National Conference on Artificial Intelligence, American Association for Artificial Intelligence (1985) 284–290
Koton, P. Empirical and model-based reasoning in expert systems. In Proceedings of the Ninth International Joint Conference on Artificial Intelligence (1985) 297–299
Koton, P. Reasoning about evidence in causal explanations. In Proceedings of the National Conference on Artificial Intelligence, American Association for Artificial Intelligence (1988)
Koton, P. Using Experience in Learning and Problem Solving. PhD thesis, Massachusetts Institute of Technology (1988)
Long, W., Naimi, S., Criscitiello, M., Jayes, R. The development and use of a causal model for reasoning about heart failure. In Symposium on Computer Applications in Medical Care, IEEE (1987) 30–36
Patil, R. Causal representation of patient illness for electrolyte and acid-base diagnosis. TR 267, Massachusetts Institute of Technology, Laboratory for Computer Science, 545 Technology Square, Cambridge, MA, 02139 (1981)
Simmons, R., Davis, R. Generate, test, and debug: combining associational rules and causal models. In Proceedings of the Tenth International Joint Conference on Artificial Intelligence (1987) 1071–1078
Simmons, R. Generate, Test, and Debug: A Paradigm for Solving Interpretation and Planning Problems. PhD thesis, Massachusetts Institute of Technology (1988)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 1993 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Koton, P.A. (1993). Combining Causal Models and Case-Based Reasoning. In: David, JM., Krivine, JP., Simmons, R. (eds) Second Generation Expert Systems. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-77927-5_4
Download citation
DOI: https://doi.org/10.1007/978-3-642-77927-5_4
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-77929-9
Online ISBN: 978-3-642-77927-5
eBook Packages: Springer Book Archive