To Select or To Weigh: A Comparative Study of Model Selection and Model Weighing for SPODE Ensembles
Y. Yang, G. Webb, J. Cerquides, K. Korb, J. Boughton, and K. Ting. Lecture Notes in Computer Science 4212: Proceedings of the 17th European Conference on Machine Learning (ECML'06), page 533-544. Berlin/Heidelberg, Springer-Verlag, (2006)
Abstract
An ensemble of Super-Parent-One-Dependence Estimators (SPODEs) offers a powerful yet simple alternative to naive Bayes classifiers, achieving significantly higher classification accuracy at a moderate cost in classification efficiency. Currently there exist two families of methodologies that ensemble candidate SPODEs for classification. One is to select only helpful SPODEs and uniformly average their probability estimates, a type of model selection. Another is to assign a weight to each SPODE and linearly combine their probability estimates, a methodology named model weighing. This paper presents a theoretical and empirical study comparing model selection and model weighing for ensembling SPODEs. The focus is on maximizing the ensemble's classification accuracy while minimizing its computational time. A number of representative selection and weighing schemes are studied, providing a comprehensive research on this topic and identifying effective schemes that provide alternative trades-offs between speed and expected error
%0 Conference Paper
%1 YangWebbCerquideKorbBoughtonTing06
%A Yang, Y.
%A Webb, G.I.
%A Cerquides, J.
%A Korb, K.
%A Boughton, J.
%A Ting, K-M.
%B Lecture Notes in Computer Science 4212: Proceedings of the 17th European Conference on Machine Learning (ECML'06)
%C Berlin/Heidelberg
%D 2006
%E Furkranz, J.
%E Scheffer, T.
%E Spiliopoulou, M.
%I Springer-Verlag
%K Conditional Estimation,AODE Probability
%P 533-544
%T To Select or To Weigh: A Comparative Study of Model Selection and Model Weighing for SPODE Ensembles
%X An ensemble of Super-Parent-One-Dependence Estimators (SPODEs) offers a powerful yet simple alternative to naive Bayes classifiers, achieving significantly higher classification accuracy at a moderate cost in classification efficiency. Currently there exist two families of methodologies that ensemble candidate SPODEs for classification. One is to select only helpful SPODEs and uniformly average their probability estimates, a type of model selection. Another is to assign a weight to each SPODE and linearly combine their probability estimates, a methodology named model weighing. This paper presents a theoretical and empirical study comparing model selection and model weighing for ensembling SPODEs. The focus is on maximizing the ensemble's classification accuracy while minimizing its computational time. A number of representative selection and weighing schemes are studied, providing a comprehensive research on this topic and identifying effective schemes that provide alternative trades-offs between speed and expected error
@inproceedings{YangWebbCerquideKorbBoughtonTing06,
abstract = {An ensemble of Super-Parent-One-Dependence Estimators (SPODEs) offers a powerful yet simple alternative to naive Bayes classifiers, achieving significantly higher classification accuracy at a moderate cost in classification efficiency. Currently there exist two families of methodologies that ensemble candidate SPODEs for classification. One is to select only helpful SPODEs and uniformly average their probability estimates, a type of model selection. Another is to assign a weight to each SPODE and linearly combine their probability estimates, a methodology named model weighing. This paper presents a theoretical and empirical study comparing model selection and model weighing for ensembling SPODEs. The focus is on maximizing the ensemble's classification accuracy while minimizing its computational time. A number of representative selection and weighing schemes are studied, providing a comprehensive research on this topic and identifying effective schemes that provide alternative trades-offs between speed and expected error},
added-at = {2016-03-20T05:42:04.000+0100},
address = {Berlin/Heidelberg},
author = {Yang, Y. and Webb, G.I. and Cerquides, J. and Korb, K. and Boughton, J. and Ting, K-M.},
biburl = {https://www.bibsonomy.org/bibtex/216dc994fa578d38ff1f4630100755d9d/giwebb},
booktitle = {Lecture Notes in Computer Science 4212: Proceedings of the 17th European Conference on Machine Learning (ECML'06)},
editor = {Furkranz, J. and Scheffer, T. and Spiliopoulou, M.},
interhash = {d9b435b54afd7f7e68d958d61bd36f35},
intrahash = {16dc994fa578d38ff1f4630100755d9d},
keywords = {Conditional Estimation,AODE Probability},
location = {Berlin, Germany},
pages = {533-544},
publisher = {Springer-Verlag},
timestamp = {2016-03-20T05:42:04.000+0100},
title = {To Select or To Weigh: A Comparative Study of Model Selection and Model Weighing for SPODE Ensembles},
year = 2006
}