Improving classification through ensemble neural networks
Khobaib Zaamout, John Z. Zhang · 2012
We consider using neural networks as an ensemble technique to improve classification accuracy. Neural networks are among the best techniques used for classification. In this work, we make use of ensemble approach to combine individual neural networks' outputs by another neural network. Furthermore, we propose to include original data as additional inputs for the ensemble neural network. The effectiveness of our proposed approach is demonstrated through a series of experiments on real and synthetic datasets.