Abstract
The use of different discretization techniques can be expected to affect the classification bias and variance of naive-Bayes classifiers. We call such an effect discretization bias and variance. Proportional k-interval discretization (PKID) tunes discretization bias and variance by adjusting discretized interval size and number proportional to the number of training instances. Theoretical analysis suggests that this is desirable for naive-Bayes classifiers. However PKID is sub-optimal when learning from training data of small size. We argue that this is because PKID equally weighs bias reduction and variance reduction. But for small data, variance reduction can contribute more to lower learning error and thus should be given greater weight than bias reduction. Accordingly we propose weighted proportional k-interval discretization (WPKID), which establishes a more suitable bias and variance trade-off for small data while allowing additional training data to be used to reduce both bias and variance. Our experiments demonstrate that for naive-Bayes classifiers, WPKID improves upon PKID for smaller datasets with significant frequency; and WPKID delivers lower classification error significantly more often than not in comparison to three other leading alternative discretization techniques studied.
’Small ‘is a relative rather than an absolute term. Of necessity, we here utilize an arbitrary definition, deeming datasets with size no larger than 1000 as ’smaller‘ datasets, otherwise as ’larger‘ datasets.
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
Bay, S. D. The UCI KDD archive [http://kdd.ics.uci.edu], 1999. Department of Information and Computer Science, University of California, Irvine.
Blake, C. L., and Merz, C. J. UCI repository of machine learning databases [http://www.ics.uci.edu/~mlearn/mlrepository.html], 1998. Department of Information and Computer Science, University of California, Irvine.
Cestnik, B. Estimating probabilities: A crucial task in machine learning. In Proc. of the European Conf. on Artificial Intelligence (1990), pp. 147–149.
Domingos, P., and Pazzani, M. On the optimality of the simple Bayesian classifier under zero-one loss. Machine Learning 29 (1997), 103–130.
Dougherty, J., Kohavi, R., and Sahami, M. Supervised and unsupervised discretization of continuous features. In Proc. of the Twelfth International Conf. on Machine Learning (1995), pp. 194–202.
Fayyad, U. M., and Irani, K. B. Multi-interval discretization of continuous-valued attributes for classification learning. In Proc. of the Thirteenth International Joint Conf. on Artificial Intelligence (1993), pp. 1022–1027.
Friedman, J. H. On bias, variance, 0/1-loss, and the curse-of-dimensionality. Data Mining and Knowledge Discovery 1,1 (1997), 55–77.
Gama, J., Torgo, L., and Soares, C. Dynamic discretization of continuous attributes. In Proc. of the Sixth Ibero-American Conf. on AI (1998), pp. 160–169.
Hsu, C. N., Huang, H. J., and Wong, T. T. Why discretization works for naive Bayesian classifiers. In Proc. of the Seventeenth International Conf. on Machine Learning (2000), pp. 309–406.
Hussain, F., Liu, H., Tan, C. L., and Dash, M. Discretization: An enabling technique, 1999. Technical Report, TRC6/99, School of Computing, National University of Singapore.
Johnson, R., and Bhattacharyya, G.Statistics: Principles and Methods. John Wiley & Sons Publisher, 1985.
Kohavi, R., and Wolpert, D. Bias plus variance decomposition for zero-one loss functions. In Proc. of the Thirteenth International Conf. on Machine Learning (1996), pp. 275–283.
Kong, E. B., and Dietterich, T. G. Error-correcting output coding corrects bias and variance. In Proc. of the Twelfth International Conf. on Machine Learning (1995), pp. 313–321.
Kononenko, I. Naive Bayesian classifier and continuous attributes. Informatica 16,1 (1992), 1–8.
Kononenko, I. Inductive and Bayesian learning in medical diagnosis. Applied Artificial Intelligence 7 (1993), 317–337.
Mora, L., Fortes, I., Morales, R., and Triguero, F. Dynamic discretization of continuous values from time series. In Proc. of the Eleventh European Conf. on Machine Learning (2000), pp. 280–291.
Pazzani, M. J. An iterative improvement approach for the discretization of numeric attributes in Bayesian classifiers. In Proc. of the First International Conf. on Knowledge Discovery and Data Mining (1995).
Quinlan, J. R.C4.5: Programs for Machine Learning. Morgan Kaufmann Publishers, 1993.
Scheaffer, R. L., and McClave, J. T.Probability and Statistics for Engineers, fourth ed. Duxbury Press, 1995.
Silverman, B.Density Estimation for Statistics and Data Analysis. Chapman and Hall Ltd., 1986.
Torgo, L., and Gama, J. Search-based class discretization. In Proc. of the Ninth European Conf. on Machine Learning (1997), pp. 266–273.
Webb, G. I. Multiboosting: A technique for combining boosting and wagging. Machine Learning 40,2 (2000), 159–196.
Weiss, N. A.Introductory Statistics, vol. Sixth Edition. Greg Tobin, 2002.
Yang, Y., and Webb, G. I. Proportional k-interval discretization for naive-Bayes classifiers. In Proc. of the Twelfth European Conf. on Machine Learning (2001), pp. 564–575.
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2003 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Yang, Y., Webb, G.I. (2003). Weighted Proportional k-Interval Discretization for Naive-Bayes Classifiers. In: Whang, KY., Jeon, J., Shim, K., Srivastava, J. (eds) Advances in Knowledge Discovery and Data Mining. PAKDD 2003. Lecture Notes in Computer Science(), vol 2637. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-36175-8_50
Download citation
DOI: https://doi.org/10.1007/3-540-36175-8_50
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-04760-5
Online ISBN: 978-3-540-36175-6
eBook Packages: Springer Book Archive