A Comparative Study of Semi-naive Bayes Methods in Classification Learning
F. Zheng, and G. Webb. Proceedings of the Fourth Australasian Data Mining Conference (AusDM05), page 141-156. Sydney, University of Technology, (2005)
Abstract
Numerous techniques have sought to improve the accuracy of Naive Bayes (NB) by alleviating the attribute interdependence problem. This paper summarizes these semi-naive Bayesian methods into two groups: those that apply conventional NB with a new attribute set, and those that alter NB by allowing inter-dependencies between attributes. We review eight typical semi-naive Bayesian learning algorithms and perform error analysis using the bias-variance decomposition on thirty-six natural domains from the UCI Machine Learning Repository. In analysing the results of these experiments we provide general recommendations for selection between methods.
%0 Conference Paper
%1 ZhengWebb05
%A Zheng, F.
%A Webb, G.I.
%B Proceedings of the Fourth Australasian Data Mining Conference (AusDM05)
%C Sydney
%D 2005
%E Simoff, S.J.
%E Williams, G.J.
%E Galloway, J.
%E Kolyshkina, I.
%I University of Technology
%K AODE, Conditional Estimation Probability
%P 141-156
%T A Comparative Study of Semi-naive Bayes Methods in Classification Learning
%X Numerous techniques have sought to improve the accuracy of Naive Bayes (NB) by alleviating the attribute interdependence problem. This paper summarizes these semi-naive Bayesian methods into two groups: those that apply conventional NB with a new attribute set, and those that alter NB by allowing inter-dependencies between attributes. We review eight typical semi-naive Bayesian learning algorithms and perform error analysis using the bias-variance decomposition on thirty-six natural domains from the UCI Machine Learning Repository. In analysing the results of these experiments we provide general recommendations for selection between methods.
@inproceedings{ZhengWebb05,
abstract = {Numerous techniques have sought to improve the accuracy of Naive Bayes (NB) by alleviating the attribute interdependence problem. This paper summarizes these semi-naive Bayesian methods into two groups: those that apply conventional NB with a new attribute set, and those that alter NB by allowing inter-dependencies between attributes. We review eight typical semi-naive Bayesian learning algorithms and perform error analysis using the bias-variance decomposition on thirty-six natural domains from the UCI Machine Learning Repository. In analysing the results of these experiments we provide general recommendations for selection between methods.},
added-at = {2016-03-20T05:42:04.000+0100},
address = {Sydney},
author = {Zheng, F. and Webb, G.I.},
biburl = {https://www.bibsonomy.org/bibtex/295dbd2981112a8454dfbb85d748230f2/giwebb},
booktitle = {Proceedings of the Fourth Australasian Data Mining Conference (AusDM05)},
editor = {Simoff, S.J. and Williams, G.J. and Galloway, J. and Kolyshkina, I.},
interhash = {50b93b858e085f20bb4f031072ed09e2},
intrahash = {95dbd2981112a8454dfbb85d748230f2},
keywords = {AODE, Conditional Estimation Probability},
location = {Sydney, Australia},
pages = {141-156},
publisher = {University of Technology},
timestamp = {2016-03-20T05:42:04.000+0100},
title = {A Comparative Study of Semi-naive Bayes Methods in Classification Learning},
year = 2005
}