Abstract
Machine learning algorithms like Genetic Programming (GP) can evolve biased classifiers when data sets are unbalanced. In this paper we compare the effectiveness of two GP classification strategies. The first uses the standard (zero) class-threshold, while the second uses the “best” class-threshold determined dynamically on a solution-by-solution basis during evolution. These two strategies are evaluated using five different GP fitness across across a range of binary class imbalance problems, and the GP approaches are compared to other popular learning algorithms, namely, Naive Bayes and Support Vector Machines. Our results suggest that there is no overall difference between the two strategies, and that both strategies can evolve good solutions in binary classification when used in combination with an effective fitness function.
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
Fawcett, T., Provost, F.: Adaptive fraud detection. Data Mining and Knowledge Discovery 1, 291–316 (1997)
Holmes, J.H.: Differential negative reinforcement improves classifier system learning rate in two-class problems with unequal base rates. In: Koza, J.R., Banzhaf, W., Chellapilla, K., et al. (eds.) Genetic Programming 1998: Proceedings of the Third Annual Conference, pp. 635–644. Morgan Kaufmann, San Francisco (1998)
Munder, S., Gavrila, D.: An experimental study on pedestrian classification. IEEE Transactions on Pattern Analysis and Machine Intelligence 28, 1863–1868 (2006)
Koza, J.R.: Genetic Programming: On the Programming of Computers by Means of Natural Selection. MIT Press, Cambridge (1992)
Winkler, S., Affenzeller, M., Wagner, S.: Advanced genetic programming based machine learning. Journal of Mathematical Modelling and Algorithms 6(3), 455–480 (2007)
Doucette, J., Heywood, M.I.: GP classification under imbalanced data sets: Active sub-sampling and AUC approximation. In: O’Neill, M., Vanneschi, L., Gustafson, S., Esparcia Alcázar, A.I., De Falco, I., Della Cioppa, A., Tarantino, E. (eds.) EuroGP 2008. LNCS, vol. 4971, pp. 266–277. Springer, Heidelberg (2008)
Bhowan, U., Johnston, M., Zhang, M.: Genetic programming for image classification with unbalanced data (ivcnz). In: Proceedings of 24th International Conference on Image and Vision Computing, Wellington, New Zealand, pp. 316–321. IEEE Press, Los Alamitos (2009)
Zhang, M., Smart, W.: Using Gaussian distribution to construct fitness functions in genetic programming for multiclass object classification. Pattern Recognition Letters 27(11), 1266–1274 (2006)
Bradley, A.P.: The use of the area under the ROC curve in the evaluation of machine learning algorithms. Pattern Recognition 30, 1145–1159 (1997)
Yan, L., Dodier, R., Mozer, M.C., Wolniewicz, R.: Optimizing classifier performance via the Wilcoxon-Mann-Whitney statistic. In: Proceedings of The Twentieth International Conference on Machine Learning (ICML 2003), pp. 848–855 (2003)
Hall, M., Frank, E., Holmes, G., Pfahringer, B., Reutemann, P., Witten, I.H.: The WEKA data mining software: An update. SIGKDD Explorations 11(1) (2009)
Asuncion, A., Newman, D.: UCI Machine Learning Repository (2007), http://www.ics.uci.edu/~mlearn/MLRepository.html , University of California, Irvine, School of Information and Computer Sciences
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2010 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Bhowan, U., Zhang, M., Johnston, M. (2010). A Comparison of Classification Strategies in Genetic Programming with Unbalanced Data. In: Li, J. (eds) AI 2010: Advances in Artificial Intelligence. AI 2010. Lecture Notes in Computer Science(), vol 6464. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-17432-2_25
Download citation
DOI: https://doi.org/10.1007/978-3-642-17432-2_25
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-17431-5
Online ISBN: 978-3-642-17432-2
eBook Packages: Computer ScienceComputer Science (R0)