Ensemble Selection for SuperParent-One-Dependence Estimators

SuperParent-One-Dependence Estimators (SPODEs) loosen Naive-Bayes’ attribute independence assumption by allowing each attribute to depend on a common single attribute (superparent) in addition to the class. An ensemble of SPODEs is able to achieve high classification accuracy with modest computational cost. This paper investigates how to select SPODEs for ensembling. Various popular model selection strategies are presented. Their learning efficacy and efficiency are theoretically analyzed and empirically verified. Accordingly, guidelines are investigated for choosing between selection criteria in differing contexts. Content areas Bayesian networks, machine learning Pre-publication draft of paper accepted for publication in the Proceedings of AI-2005 which will be published as Springer LNAI 

• Type: Conference
• Year: 2005
• Where: AUSAI
• Authors: Ying Yang, Kevin B. Korb, Kai Ming Ting, Geoffrey I. Webb
• Keywords: Artificial Intelligence, Attribute Independence Assumption, AUSAI 2005, Common Single Attribute, Modest Computational Cost