Abstract
Two issues are central to effective case-based reasoning: deriving good indices for retrieving cases and effectively using those cases. A number of straightforward methods for using cases, such as interpolation, can provide good results. However, effective indexing strategies have proven more elusive. This paper presents a new and effective approach based on qualitative modelling. We build a partial or complete qualitative model of a physical system, and use this model to derive the minimal sets of parameters relevant to each of the desired inputs. This reduces the number of attributes used for indexing, thereby increasing the density of cases in the parameter space. We present results showing the effectiveness of the approach on a simplified sewage treatment plant.
This work has been funded jointly by Swiss National Science Foundation grant #5003-034269 and by Nestle York RECO, England. Particular thanks to Prof. Boi Faltings (EPFL) and Mr. Peter Duxbury-Smith (Nestle) for their help and inspiration.
This is a preview of subscription content, log in via an institution to check access.
Preview
Unable to display preview. Download preview PDF.
References
D. W. Aha, D. Kibler, and M. K. Albert: Instance-based learning algorithms. Machine Learning 6 (1991) pp 37–66.
C. Cardie: Using decision trees to improve case-based learning. Proceedings of the Tenth International Conference on Machine Learning (1993) pp 25–32.
T. H. Cover and P. E. Hart: Nearest neighbor pattern classification. IEEE Transactions on Information Theory 13 (1967) pp 21–47.
M. Goodman: Results on controlling action with projective visualization. Proceedings of the Twelfth National Conference on Artificial Intelligence (1994) pp 1245–1250.
J. J. Parry and C. X. Ling: Deciding Weights for IBL Algorithms using C4.5. unpublished draft (1994).
J. R. Quinlan: C4.5: Programs for Machine Learning 1 (1992).
J. R. Quinlan: Combining instance-based and model-based learning. Proceedings of the Tenth International Conference on Machine Learning (1993) pp 236–243.
S. Ramachandran, R. Mooney, and B. Kuipers: Learning Qualitative Models for Systems with Multiple Operating Regions. Proceedings of the Eighth International Workshop on Qualitative Reasoning about Physical Systems (1994) pp 212–223.
B. L. Richards, I. Kraan, and B. Kuipers: Automatic abduction of qualitative models. Proceedings of the Tenth National Conference on Artificial Intelligence (1992) pp 723–728.
M. M. Veloso: Planning and Learning by Analogical Reasoning, Springer-Verlag Lecture Notes in Artificial Intelligence (1994).
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1995 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Richards, B.L. (1995). Qualitative models as a basis for case indices. In: Haton, JP., Keane, M., Manago, M. (eds) Advances in Case-Based Reasoning. EWCBR 1994. Lecture Notes in Computer Science, vol 984. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-60364-6_32
Download citation
DOI: https://doi.org/10.1007/3-540-60364-6_32
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-60364-1
Online ISBN: 978-3-540-45052-8
eBook Packages: Springer Book Archive