Abstract
This paper describes the use of an evolutionary algorithm to develop lookup tables which consist of an ordered list of regions, each of which encloses training examples of only one category. Compared to a simpler type of lookup table which consists of an unordered list of the training points and their categories, region based tables are smaller and, in general, faster to use. The development of a region based lookup table for the Frey and Slate character recognition problem is described and the size and accuracy achieved are compared with the original Frey and Slate point based lookup table. The reasons why it outperforms the original lookup table are discussed.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsPreview
Unable to display preview. Download preview PDF.
References
John R Podlena and Tim Hendtlass: Using a combination of Genetic Algorithm and Hill Climbing Techniques for the solution of a difficult letter recognition problem. Proc. Of the Eighth IEA/AEI International Conference, 1995 (55–60)
Carlos Fonseca and Peter Flemming: An overview of Evolutionary Algorithms in Multiobjective Optimisation. Evolutionary Computing, volume 3, number 1. 1995.(1–16)
P.W.Frey and D.J.Slate: Letter recognition using Holland Style Adaptive Classifiers. Machine Learning, volume 6 number 2, (1991).
G Jones: Letter Image Recognition Data www.sgi.conVTechnology/mlc/db/letter.names
D. Philpot, personal communication.
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1998 Springer-Verlag
About this paper
Cite this paper
Hendtlass, T. (1998). Generating lookup tables using evolutionary algorithms. In: Mira, J., del Pobil, A.P., Ali, M. (eds) Methodology and Tools in Knowledge-Based Systems. IEA/AIE 1998. Lecture Notes in Computer Science, vol 1415. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-64582-9_796
Download citation
DOI: https://doi.org/10.1007/3-540-64582-9_796
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-64582-5
Online ISBN: 978-3-540-69348-2
eBook Packages: Springer Book Archive