Improved MLP neural network as chromosome classifier

In this paper we introduce a technique to reduce dimension of Neural Networks (NN) for classification and apply it to a Multi Layer Perceptron (MLP) NN. This technique reduces number of output neurons from an order of n to log₂{n} that reduces dimension of network, number of required training data, generalization error of the network and training time significantly. The proposed technique is employed for human chromosome classification using Copenhagen data set. Using 304 chromosomes for 24 classes in training mode, a faster training time in compare to standard MLP and accuracy more than 88% in recall mode is achieved. The introduced idea can be generalized to any Neural Network, which is used for classification.