Induction of Selective Bayesian Network Classiiers

We present an algorithm for inducing Bayesian networks using feature selection. The algorithm selects a subset of attributes that maximizes predictive accuracy prior to the network learning phase, thereby incorporating a bias for small networks that retain high predictive accuracy. We compare the behavior of this selective Bayesian network classiier with that of (a) Bayesian network classiiers that incorporate all attributes, (b) selective and non-selective naive Bayesian classiiers, and (c) the decision-tree algorithm C4.5. With respect to (a), we show that our approach generates networks that are computationally simpler to evaluate but display comparable predictive accuracy. With respect to (b), we show that the selective Bayesian network classiier performs signiicantly better than both versions of the naive Bayesian classiier on almost all databases studied, and hence is an enhancement of the naive method. With respect to (c), we show that the selective Bayesian network classiier displays comparable behavior.