We propose the Dirichlet Process - Naive Bayes model (DPNB) that can simultaneously impute missing values and address classification problems.
We have been studying the theoretical properties of Bayesian neural network (BNN) models and their applications.
