Comparing Bayesian Network Classifiers

TL;DR

This paper empirically compares various Bayesian network classifiers, demonstrating that CI-based learning algorithms produce competitive classifiers with efficient computation, and proposing a new algorithm that further improves classifier effectiveness.

Contribution

It introduces and evaluates CI-based algorithms for learning BN classifiers, showing their competitiveness and efficiency, and proposes a new algorithm that enhances classifier performance.

Findings

CI-based BN classifiers are competitive with state-of-the-art methods.

Learning and inference with these classifiers are computationally efficient.

A new algorithm further improves classifier effectiveness.

Abstract

In this paper, we empirically evaluate algorithms for learning four types of Bayesian network (BN) classifiers - Naive-Bayes, tree augmented Naive-Bayes, BN augmented Naive-Bayes and general BNs, where the latter two are learned using two variants of a conditional-independence (CI) based BN-learning algorithm. Experimental results show the obtained classifiers, learned using the CI based algorithms, are competitive with (or superior to) the best known classifiers, based on both Bayesian networks and other formalisms; and that the computational time for learning and using these classifiers is relatively small. Moreover, these results also suggest a way to learn yet more effective classifiers; we demonstrate empirically that this new algorithm does work as expected. Collectively, these results argue that BN classifiers deserve more attention in machine learning and data mining communities.