Bayesian Network Inference Based on Functional Dependency Mining of Relational D

In order to construct robust and flexible Bayesian network, this paper proposed functional dependency rules of probability to create candidate key and delete extraneous attributes. The functional dependencies implicated in each sample will be found based on association rule mining technique in the context of classification. The corresponding learning algorithm, namely FDBC (Functional Dependency based Bayesian network Classifier), relaxes the assumption of conditional independence while maintaining inter-dependencies between attributes. Experimental results are presented to show the effectiveness and efficiency of the proposed approach.