Abstract
This paper presents a novel Genetic Algorithm (GA) [1], [2] based approach for decision tree induction. Decision tree induction algorithms such as ID3 [3], [4] and CHAID [5], are based on a stepwise search procedure. This is essentially a heuristic search technique based on selecting the best local attribute/values split for each internal node. This is performed according to some given criteria, regardless of the impact on subsequent splits. Once a split is selected, these algorithms have no backtracking mechanism to enable them to change an attribute split. Thus a potential weakness of these algorithms is the lack of a globally optimal search strategy.
GAs perform a non-linear search for the optimal or near optimal solution in a pre-defined search space. The aim is to cover the entire search space without performing an exhaustive search of the domain. This paper presents GAs as an effective alternative to the step-wise search strategy employed by traditional decision tree induction algorithms. The new algorithm has been applied to two data sets and has shown a clear improvement in classification accuracy over ID3 generated by a more traditional methods.
This is a preview of subscription content, log in via an institution to check access.
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
Holland, J.: Adaptation in Natural and Artificial Systems. University of Michigan (1975).
Grefenstett, J.: Genetic Algorithms For Machine Learning. A Special Issue of Machine Learning. Machine Learning, Vol. 13, Nos 2-3. Kluwer Academic Publishers (1992).
Quinlan, J.: Induction of Decision Trees. Machine Learning 1, Kluwer Academic Press (1986).
Michie, D.: Personal Models of Rationality. Technical Report, Turing Institute Document No. TIRM-88-035 (1989).
Kass, G. V.: An exploratory Technique for Investigating Large Quantities of Categorical Data. Applied statistics, 29, No. 2 (1980).
Quinlan, J.: Simplifying Decision Trees. International Journal of Man-Machine studies (1987).
Koza, J.: Concept formation and Decision Tree Induction Using the Genetic Peogramming Paradigm. Proceeding of the 1st Workshop. Parallel Problem Solving from Nature. Dortmund, Germany. Springer-Verlag (1991).
Goldberg, D.: A Meditation on the Application of Genetic Algorithms. IlliGal Technical Report No. 98003 (1998).
Bratko, I., and Kononenko, I.: Learning Diagnostic Rules from Incomplete and Noisy Data. Technical Report. Turing Institute Document No.ANT24362(1987).
Michie D., Spiegelhalter D.J., Taylor C.C.: Machine Learning, Neural and Statistical Classification, Ellis Hopwood Series in Artificial Intelligence, Ellis Hopwood, (1994).
Davis, L.: Handbook of Genetic Algorithms. Van Nostrand, Rheinhold (1991).
Author information
Authors and Affiliations
Rights and permissions
Copyright information
© 1999 Springer-Verlag Wien
About this paper
Cite this paper
Bandar, Z., Al-Attar, H., Crockett, K. (1999). Genetic Algorithms For Decision Tree Induction. In: Artificial Neural Nets and Genetic Algorithms. Springer, Vienna. https://doi.org/10.1007/978-3-7091-6384-9_32
Download citation
DOI: https://doi.org/10.1007/978-3-7091-6384-9_32
Publisher Name: Springer, Vienna
Print ISBN: 978-3-211-83364-3
Online ISBN: 978-3-7091-6384-9
eBook Packages: Springer Book Archive