Learning Bayesian Networks: A Unification for Discrete and Gaussian Domains
This work provides a unified framework for Bayesian network learning, which is incremental as it builds on previous conference presentations.
The authors tackled the problem of learning Bayesian networks from prior knowledge and statistical data by unifying approaches for discrete and Gaussian domains, resulting in a general Bayesian scoring metric applicable to both domains.
We examine Bayesian methods for learning Bayesian networks from a combination of prior knowledge and statistical data. In particular, we unify the approaches we presented at last year's conference for discrete and Gaussian domains. We derive a general Bayesian scoring metric, appropriate for both domains. We then use this metric in combination with well-known statistical facts about the Dirichlet and normal--Wishart distributions to derive our metrics for discrete and Gaussian domains.