Competence-Guided Case-Base Editing
Techniques

Elizabeth McKenna and Barry Smyth

Smart Media Institute, Department of Computer Science
University College Dublin, Belfield, Dublin 4, Ireland
{Elizabeth.McKenna,Barry.Smyth}@ucd.ie



Abstract. Case-based classification is a powerful classification method,
which (in its simplest form) assigns a target case to the same class as
the nearest of n previously classified cases. Many case-based classifiers
use the simple nearest-neighbour technique to identify the nearest case,
but this means comparing the target case to all of the stored cases at
classification time, resulting in high classification costs. For this reason
many techniques have been proposed to improve the performance of case-based 
classifiers by reducing the search they must perform. In this paper
we will look at editing techniques that preserve the lazy-learning quality
of case-based classification, hut improve classification performance.
References

1.	Aha, D. W., Kibler, D., and Albert, M. K.: Instance-Based Learning Algorithms,
Machine Learning 6 (1991), 3766.
2.	Blake, C., Keogh, E., and Merz, C. J.: UCI repository of Machine Learning
Databases, Irvine, CA: University of California. Department of Information and
Computer Science, (1998).
3.	Brighton, H.: Information Filtering for Lazy Learning Algorithms, Masters thesis,
Centre for Cognitive Science, University of Edinburgh, Scotland, (1997).
4.	Brighton, H and Mellish, C.: On the Consistency of information filters for Lazy
learning algorithms, Proceedings of the 3rd European Conference on the Principles
of Data Mining and Knowledge Discovery. Lecture Notes in Computer Science
(Jan Rauch Jan M. Zytkow, ed.), Springer Verlag, (1999), pp. 283288.
5.	Cameron-Jones, R. M.: Minimum Description Length Case-Based Learning., Proceedings 
of the 5th Australian Joint Conference on Artificial Intelligence, World
Scientific, (1992), pp. 368373.
6.	Chang, C. L.: Finding Prototypes for Nearest Neighbour Classifiers, IEEE Transactions 
on Computers C-23 (1974), 11791184.
7.	Dasarathy, D. V.: (ed.), Nearest Neighbor Norms: NN Pattern Classification Techniques, 
IEEE Press, (1991).
8.	Cates, C. W.: The Reduced Nearest Neighbor Rule, IEEE Transactions on Information 
Theory IT-18(3) (1972), 431433.
9.	Hart, P. E.: The Condensed Nearest Neighbor Rule, IEEE Transactions on Information 
Theory IT-14 (1967), 515516.
10.	King, R. D., Feng, C. and Sutherland, A.: Statlog: Comparison of Classification
Algorithms on Large Real- Worlds Problems, Applied Artificial Intelligence 9(3)
(1995), 289333.
11.	McKenna, E. and Smyth, B.: Competence-guided Editing Methods for Lazy Learning, 
Proceedings of the 14th European Conference on Artificial Intelligence, (2000).
12.	Smyth, B. and Keane, M. T.: Remembering to Forget: A Competence Preserving 
Case Deletion Policy for CBR Systems, Proceedings of the 14th International
Joint Conference on Artificial Intelligence (Chris Mellish, ed.), Morgan Kaufmann
(1995), pp. 377382.
13.	Smyth, B. and McKenna, E.: Modelling the Competence of Case-Bases, Advances
in Case-Based Reasoning. Lecture Notes in Artificial Intelligence (B. Smyth and
P. Cunningharn, eds.), Springer Verlag, (1998), pp. 208220.
14.	Smyth, B. and McKenna, E.: Building Compact Competent Case-Bases, Case-Based 
Reasoning Research and Development. Lecture Notes in Artificial Intelligence 
(Klaus Dieter Althoff, Ralph Bergmann, and L.Karl Branting, ads.), Springer
Verlag, (1999), pp. 329342.
15.	Smyth, B. and McKenna, E.: Footprint-Based Retrieval, Case-Based Reasoning
Research and Development. Lecture Notes in Artificial Intelligence (Klaus Dieter
Althoff, Ralph Bergmann, and L.Karl Branting, eds.), Springer Verlag, (1999),
pp. 343357.
16.	Smyth, B. and McKenna, E.: An Efficient and Effective Procedure for Updating a
Competence Model for Case-Based Reasoners, Proceedings of the 11th European
Conference on Machine Learning (Ramon Lopez de Mantaras and Enric Plaza
eds.), Springer Verlag, (2000).
17.	Tomek, I.: Two Modifications of CNN, IEEE Transactions on Systems, Man, and
Cybernetics 7(2) (1976), 679772.
18.	Wilson, D. L.: Asymptotic Properties of Nearest Neighbor Rules Using Edited Data,
IEEE Transactions on Systems, Man, and Cybernetics 2-3 (1972), 408421.
19.	Wilson, D. R. and Martinez, T. R.: Instance Pruning Techniques, Proceedings of
the 14th International Conference on Machine Learning (D. Fisher, ed.), Morgan
Kaufmann, (1997), pp. 403411.
