A Constructive Algorithm for Training Neural Network Ensemble

Neural network ensemble is a learning paradigm where many neural networks are used together to solve a particular problem. This paper presents a new method to construct a neural network ensemble (NNE) based on Correlation, Interaction Validation and Entropy (CIENNE). The method consists of two parts: a sub-algorithm to construct best component neural networks with Correlation and Interaction Validation, and a sub-algorithm to combine the component neural networks with Entropy. Experimental results demonstrate that the proposed approach is effective.