A novel modular neural network for imbalanced classification problems

In this paper, a novel modular neural network is proposed to solve multi-class problems with imbalanced training sets. The proposed model can transform an imbalanced classification problem into a set of symmetrical two-class problems, each of which is solved by single neural network with a simple structure. The results of all neural networks are then combined by averaging or GA method to form a final classification decision. The experimental results show that the proposed method reduces the time consumption for training and improves the classification performance.