A Comparative Study of Bandwidth Choice in Kernel Density Estimation for Naive Bayesian Classification

Authors:
Bin Liu;Ying Yang;Geoffrey I. Webb;Janice Boughton
Affiliations:
Clayton School of Information Technology, Monash University, Australia;Clayton School of Information Technology, Monash University, Australia;Clayton School of Information Technology, Monash University, Australia;Clayton School of Information Technology, Monash University, Australia
Venue:
PAKDD '09 Proceedings of the 13th Pacific-Asia Conference on Advances in Knowledge Discovery and Data Mining
Year:
2009

Citing 7
Cited 1

MultiBoosting: A Technique for Combining Boosting and Wagging

Machine Learning
Multivariate discretization for set mining

Knowledge and Information Systems
On Bias, Variance, 0/1—Loss, and the Curse-of-Dimensionality

Data Mining and Knowledge Discovery
Data Mining: Practical Machine Learning Tools and Techniques, Second Edition (Morgan Kaufmann Series in Data Management Systems)

Data Mining: Practical Machine Learning Tools and Techniques, Second Edition (Morgan Kaufmann Series in Data Management Systems)
Statistical Comparisons of Classifiers over Multiple Data Sets

The Journal of Machine Learning Research
Discretization for naive-Bayes learning: managing discretization bias and variance

Machine Learning
Estimating continuous distributions in Bayesian classifiers

UAI'95 Proceedings of the Eleventh conference on Uncertainty in artificial intelligence

Bayesian classifiers based on probability density estimation and their applications to simultaneous fault diagnosis

Information Sciences: an International Journal

Quantified Score

Hi-index	0.00

Visualization

Abstract

Kernel density estimation (KDE) is an important method in nonparametric learning. While KDE has been studied extensively in the context of accuracy of distribution estimation, it has not been studied extensively in the context of classification. This paper studies nine bandwidth selection schemes for kernel density estimation in Naive Bayesian classification context, using 52 machine learning benchmark datasets. The contributions of this paper are threefold. First, it shows that some commonly used and very sophisticated bandwidth selection schemes do not give good performance in Naive Bayes. Surprisingly, some very simple bandwidth selection schemes give statistically significantly better performance. Second, it shows that kernel density estimation can achieve statistically significantly better classification performance than a commonly used discretization method in Naive Bayes, but only when appropriate bandwidth selection schemes are applied. Third, this study gives bandwidth distribution patterns for the investigated bandwidth selection schemes.