Hierarchical Dependency Constrained Tree Augmented Naive Bayes Classifiers for Hierarchical Feature Spaces

TL;DR

This paper introduces two hierarchical dependency-based Tree Augmented Naive Bayes algorithms, Hie-TAN and Hie-TAN-Lite, which leverage feature hierarchies to improve classification accuracy in hierarchical feature spaces.

Contribution

The paper proposes novel hierarchical dependency-constrained TAN algorithms that incorporate feature hierarchy constraints and redundancy elimination for enhanced predictive performance.

Findings

Hie-TAN outperforms other hierarchical dependency constrained classifiers.

Hie-TAN-Lite further improves accuracy by removing hierarchical redundancy.

Experimental results demonstrate the effectiveness of the proposed methods.

Abstract

The Tree Augmented Naive Bayes (TAN) classifier is a type of probabilistic graphical model that constructs a single-parent dependency tree to estimate the distribution of the data. In this work, we propose two novel Hierarchical dependency-based Tree Augmented Naive Bayes algorithms, i.e. Hie-TAN and Hie-TAN-Lite. Both methods exploit the pre-defined parent-child (generalisation-specialisation) relationships between features as a type of constraint to learn the tree representation of dependencies among features, whilst the latter further eliminates the hierarchical redundancy during the classifier learning stage. The experimental results showed that Hie-TAN successfully obtained better predictive performance than several other hierarchical dependency constrained classification algorithms, and its predictive performance was further improved by eliminating the hierarchical redundancy, as suggested by the higher accuracy obtained by Hie-TAN-Lite.