Neural-network-based metamodeling for parallel data mining

In this study, we discuss the use of supervised neural networks as a metamodeling technique to design a practical and powerful parallel data mining system. First of all, a bagging sampling technique is used to generate different training sets. Based on the different training sets, the different neural network models with various training sets are then trained to formulate different data mining models, i.e., base mining models. Finally, a metamodel can be produced by learning from all base mining models with neural-network-based metalearning strategy. For verification, a practical data mining experiment is performed for illustration purpose.