Generation of Attribute Value Taxonomies from Data and Their Use in Data-Driven Construction of Accurate and Compact Naive Bayes Classifiers

Attribute Value Taxonomies (AVT) have been shown to be useful in constructing compact and robust classifiers. However, in many application domains, human-designed AVTs are unavailable. For this problem, we introduce AVT-Learner, an algorithm for automated construction of attribute value taxonomies from data. AVT-Learner uses Hierarchical Agglomerative Clustering (HAC) to cluster attribute values based on the distribution of classes that cooccur with the values. We describe experiments of AVT-Learner on several benchmark data sets that compare the performance of AVT-NBL (an AVT-guided Naive Bayes Learner) with that of the standard Naive Bayes Learner (NBL) applied to the original data set as well as a data set generated by augmenting the original data set with a set of additional attributes corresponding to the nodes in the AVTs. Our results show that the AVTs generated by AVT-Learner are competitive with human-generated AVTs (in cases where such AVTs are available). AVT-NBL using AVTs generated by AVT-Learner achieves classification accuracies that are comparable to or higher than that obtained by NBL; and the resulting classifiers are significantly more compact than those generated by NBL.