Using Introspective Learning to Improve Retrieval in
CBR: A Case Study in Air Traffic Control *

Andrea Bonzano1, Pdraig Cunningham1 and Barry Smyth2

1Artificial Intelligence Group, Trinity College Dublin, Ireland
2University College Dublin, Ireland

{Andrea.Bonzano,Padraig.Cunningham}@tcd.ie, bsmyth@cslan.ucd.ie


Abstract. We can learn a lot about what features are important for
retrieval by comparing similar cases in a case-base. We can determine
which features are important in predicting outcomes and we can assign
weights to features accordingly. In the same manner we can discover
which features are important in specific contexts and determine
localised feature weights that are specific to individual cases. In this
paper we describe a comprehensive set of techniques for learning local
feature weights and we evaluate these techniques on a case-base for
conflict resolution in air traffic control. We show how introspective
learning of feature weights improves retrieval and how it can be used to
determine context sensitive local weights. We also show that
introspective learning does not work well in case-bases containing only
pivotal cases because there is no redundancy to be exploited.
References
	Birnbaum, L., Collins, G., Brand, M., Freed, M., Krulwich, B., and Prior, L.
(1991) A Model-Based Approach to the Construction of Adaptive Case-Based
Planning Systems. Proceedings of the Case-Based Reasoning Workshop, pp. 2 15-224.
Washington D.C., USA.
	Bonzano A., & Cuningham P., (1996) ISAC: A CBR System for Decision
support in Air Traffic Control in Proceedings of EWCBR 96, Advances in Case-Based 
Reasoning, Ian Smith & Boi Faltings eds. Springer Verlag Lecture Notes in Al,
pp44-57.
	Fox, S. & Leake, D. B. (1995) Using Introspective Reasoning to Refine
Indexing. Proceedings of the 14th International Joint Conference on Artificial
Intelligence, pp. 391-397.
	Laird, J. E, Rosenbloom, P. 5., and Newell, A. (1986) Chucking in Soar: The
Anatomy of a General Learning Mechanism. Machine Learning. 1(1).
	Laird, J. E., Newell, A., and Rosenbloom, P. 5. (1987) Soar: An Architecture for
General Intelligence. Artificial Intelligence. 33(1).
	Leake, D. B., Kinley, A., and Wilson, D. (1995) Learning to Improve Case
Adaptation by Introspective Reasoning and CBR. Case-Based Reasoning Research
and Development (Ed.s M. Veloso & A. Aamodt), Proceedings of the 1st International
Conference on Case-Based Reasoning, pp. 229-240, Springer-Verlag.
	Munoz-Avila, H., Hllen, J. (1996) Feature Weighting by ExplainingCase-Based
Planning Episodes. Advances in Case-Based Reasoning (Ed.s I. Smith & B. Faltings),
Proceedings of the Third European Workshop on Case-Based Reasoning, pp. 280-
294, Springer-Verlag.
	Oehlmann, R., Edwards, P., & Sleeman, D. (1995) Changing the Viewpoint: Re-Indexing 
by Introspective Question. Proceedings of the 16th Annual Conference of the
Cognitive Science Society, pp. 38 1-386. Lawrence-Erlbaum and Associates.
	Saltzburg, S. L. (1991) A Nearest Hyperrectangle Learning Method. Machine
Learning, 1.
	Stefik, M. (1981) Planning and Meta-Planning. Artificial Intelligence, 16, pp.
141-170.
	Smyth, B. & Keane, M. T. (1995) Remembering to Forget: A Competence
Preserving Case Deletion Policy for CBR Systems. Proceedings of the 14th
International Joint Conference on artificial Intelligence (IJCAI-95), pp. 377-382.
Montreal, Canada.
	Veloso, M. (1992) Learning by Analogical Reasoning in General Problem
Solving. Ph.D. Thesis, (CMU-CS-92-174), School of Computer Science, Carnegie
Mellon University, Pittsburgh, USA.
	Watson I. D., (1996) Case-Based Reasoning Tools: An Overview, In
Proceedings of 2nd. UK CBR Workshop, Progress in Case-Based Reasoning, Watson.
I.D. (Ed.) pp.71-88. University of Salford. (also available on the Web at
http://l46.87.176.38/ai-cbr/Papers/cbrtools.doc)
	Wetterschereck, D., & Aha, D. W. (1995) Weighting Features. Case-Based
Reasoning Research and Development (Ed.s M. Veloso & A. Aamodt), Proceedings
of The 1st International Conference on Case-Based Reasoning, pp. 347-358,
Springer-Verlag.
	Wettschereck, D., Aba, D. W., & Mohri, T. (1997). A review and empirical
evaluation of feature weighting methods for a class of lazy learning algorithms. To
appear in Artificial Intelligence Review. (also available on the Web from
http://www.aic.nrl .navy.mil/~aha/)
