I built a BN network using the GeNIe software but for the validation part I get the same set of answers in the confusion matrix though I changed the inference algorithms and number of intervals for the discretized continuous data). Have anyone encountered such issue?
I think the problem could be the type of nodes or the learning parameters part but I could not find much information concerning the node types and learning parameters, and the documentation and tutorials which are provided by the software developer does not explain much.