%0 Conference Paper %1 YangWebb02a %A Yang, Y. %A Webb, G. I. %B Proceedings of the 2002 Pacific Rim Knowledge Acquisition Workshop (PKAW'02) %C Tokyo %D 2002 %E Yamaguchi, T. %E Hoffmann, A. %E Motoda, H. %E Compton, P. %I Japanese Society for Artificial Intelligence %K Bayes Discretization Naive for %P 159-173 %T A Comparative Study of Discretization Methods for Naive-Bayes Classifiers %X Discretization is a popular approach to handling numeric attributes in machine learning. We argue that the requirements for effective discretization differ between naive-Bayes learning and many other learning algorithms. We evaluate the effectiveness with naive-Bayes classifiers of nine discretization methods, equal width discretization (EWD), equal frequency discretization (EFD), fuzzy discretization (FD), entropy minimization discretization (EMD), iterative discretization (ID), proportional k-interval discretization (PKID), lazy discretization (LD), non-disjoint discretization (NDD) and weighted proportional k-interval discretization (WPKID). It is found that in general naive-Bayes classifiers trained on data preprocessed by LD, NDD or WPKID achieve lower classification error than those trained on data preprocessed by the other discretization methods. But LD can not scale to large data. This study leads to a new discretization method, weighted non-disjoint discretization (WNDD) that combines WPKID and NDD's advantages. Our experiments show that among all the rival discretization methods, WNDD best helps naive-Bayes classifiers reduce average classification error.

@inproceedings{YangWebb02a, abstract = {Discretization is a popular approach to handling numeric attributes in machine learning. We argue that the requirements for effective discretization differ between naive-Bayes learning and many other learning algorithms. We evaluate the effectiveness with naive-Bayes classifiers of nine discretization methods, equal width discretization (EWD), equal frequency discretization (EFD), fuzzy discretization (FD), entropy minimization discretization (EMD), iterative discretization (ID), proportional k-interval discretization (PKID), lazy discretization (LD), non-disjoint discretization (NDD) and weighted proportional k-interval discretization (WPKID). It is found that in general naive-Bayes classifiers trained on data preprocessed by LD, NDD or WPKID achieve lower classification error than those trained on data preprocessed by the other discretization methods. But LD can not scale to large data. This study leads to a new discretization method, weighted non-disjoint discretization (WNDD) that combines WPKID and NDD's advantages. Our experiments show that among all the rival discretization methods, WNDD best helps naive-Bayes classifiers reduce average classification error.}, added-at = {2016-03-20T05:42:04.000+0100}, address = {Tokyo}, audit-trail = {*}, author = {Yang, Y. and Webb, G. I.}, biburl = {https://www.bibsonomy.org/bibtex/2c6bc12b8db5bbc87da8772f02125edb1/giwebb}, booktitle = {Proceedings of the 2002 {Pacific} Rim Knowledge Acquisition Workshop (PKAW'02)}, editor = {Yamaguchi, T. and Hoffmann, A. and Motoda, H. and Compton, P.}, interhash = {db53f913a61457648e4ca5577b73c990}, intrahash = {c6bc12b8db5bbc87da8772f02125edb1}, keywords = {Bayes Discretization Naive for}, location = {Tokyo, Japan}, pages = {159-173}, publisher = {Japanese Society for Artificial Intelligence}, timestamp = {2016-03-20T05:42:04.000+0100}, title = {A Comparative Study of Discretization Methods for Naive-Bayes Classifiers}, year = 2002 }