A Neural Network For Seismic Events Classification

This paper deals with the classification problem of earthquakes by using neural networks. It is shown that such devices allow to solve the problem of automatic discrimination with high degree of accuracy even if a small number of llexamplesll are used in the learning phase of the network. The criteria used to find a optimal network configuration and to carry out the training phase are discussed. In the paper some expedients to speed-up the training of the neural network are outlined.