Determining optimal neural network architecture using regression methods

This paper presents regression methods used to determine the optimal neural network architecture. Using the results obtained from clustering the input data as independent variables to define the regression function. This function will determine the optimal number of hidden layers and the optimal number of neurons on each hidden layer for a multi-layer neural network. Data mining techniques are used to explore input data of a neural network to identify patterns and establish relationships that can lead to reasonable predictions of the optimal architecture of a multi-layer neural network.