Learning the Structure of the Tree and Tree Augmented Naive Bayesian from Incomplete and Imbalanced Data

摘要

In many fields, such as decision, diagnosis, or predication, the Bayesian network formalism is becoming increasingly common, that especially thanks particularly to its inferential capabilities, even when data is incomplete and imbalanced. In this paper, we present an approach to learn the structure of Chow-Liu and Tree Augmented Naïve Bayes (TAN) from incomplete and imbalanced datasets.