Study about Statistical Normalization and Back Propagation

Abstract

The Artificial Neural Network is one of the popular machine learning method, which has recently been applied in many areas of medical and medically related fields. Diabetes is one of the chronic diseases that occur when the blood sugar is too high. Early prediction of the disease may reduce the complications. Artificial Neural Network is known as an excellent classifier for nonlinear data. Some major issues are to be considered while constructing the network model.  The network structure, learning rate parameter and normalization of input vectors. The proposed research focuses on various normalization methods applied in back propagation neural networks to enhance the reliability of the trained network. The experimental results show that the performance of the classifier model can be increased based on the selection of the normalization method.