A good paper to read on this is often "**Bayesian Network Classifiers, Machine Learning, 29, 131–163 (1997)**".

Bayesian network models relationships between features in a very general way. If you know what these relationships are, or have enough data to derive them, then it may be appropriate to use a Bayesian network.

A Naive Bayes classifier may be an easy model that describes an explicit class of Bayesian network - wherever all of the options are class-conditionally independent. Because of this, there are bound problems that Naive Bayes cannot solve (example below). However, its simplicity also makes it easier to apply, and it requires fewer data to get a good result in many cases.

For this undergoing through the Probabilistic Bayesian Models Training would be abetter prospect for a newbie. Since they are somewhat related to Machine Learning Algorithms and also Data Science Course, one can study them too for a better grasp on the aforementioned topic.