Geoff Webb, Clayton School of Information Technology, Monash University
Abstract:
Averaged One-Dependence Estimators (AODE) relaxes the Naive Bayes (NB) independence assumption while retaining many of Naive Bayes' desirable computational and theoretical properties. This paper explores the strategy of further relaxing the independence assumption by extending the underlying strategy of AODE to higher levels of dependence. This leads to the specification of Averaged N-Dependence Estimators ($\ANDE^{n}$). We provide theoretical and experimental evidence that increasing the dependence in $\ANDE^{n}$ decreases bias at the cost of an increase in variance and computation.Extensive experimental evaluation shows that the bias-variance trade-off for Averaged 2-Dependence Estimators results in strong predictive accuracy over a wide range of data sets. We show that the asymptotic error of the highest-order variant of $\ANDE^{n}$ is the Bayes optimal error.
Speaker biographies:
Geoff holds a chair in the Faculty of Information Technology
at Monash University, where he heads the Centre for Research in Intelligent Systems.Prior to Monash he held appointments at Griffith University and then Deakin University, where he received a personal chair.His primary research areas are machine learning, data mining, and user modelling. He is known for his contribution to the debate about the application of Occam's
razor in machine learning and for the development of numerous methods, algorithms andtechniques for machine learning, data mining and user modelling. His commercial data mining software, Magnum Opus, incorporates many techniques from his association discovery research. Many of his learning algorithms are
included in the widely-used Weka machine learning workbench.He is editor-in-chief of the highest impact data mining journal,
Data Mining and Knowledge Discovery, co-editor of the Encyclopedia of Machine Learning (to be published by Springer), a member of the advisory board of Statistical Analysis and Data Mining and a member of the editorial boards of Machine Learning and ACM Transactions on Knowledge Discovery in Data.
