Interpretation of hidden node methodology with network accuracy

Abstract

Bayesian networks are constructed under a con-ditional independency assumption. This assump-tion however does not necessarily hold in prac-tice and may lead to loss of accuracy. We previ-ously proposed a hidden node methodology whereby Bayesian networks are adapted by the addition of hidden nodes to model the data de-pendencies more accurately. Empirical results in a computer vision application to classify and count the neural cell automatically showed that a modified network with two hidden nodes achieved significantly better performance with an average prediction accuracy of 83.9% com-pared to 59.31% achieved by the original net-work. In this paper we justify the improvement of performance by examining the changes in network accuracy using four network accuracy measurements; the Euclidean accuracy, the Co-sine accuracy, the Jensen-Shannon accuracy and the MDL score. Our results consistently show that the network accuracy improves by introduc-ing hidden nodes. Consequently, we were able to verify that the hidden node methodology helps to improve network accuracy and contribute to the improvement of prediction accuracy.