Estimating the relationship between isoseismal area and earthquake magnitude by a hybrid fuzzy-neural-network method

Utilizing information diffusion method and artificial neural networks, we propose in this paper a hybrid fuzzy neural network to estimate the relationship between isoseismal area and earthquake magnitude. We focus on the study of incompleteness and contradictory nature of patterns in scanty historical earthquake records. Information diffusion method is employed to construct fuzzy relationships which are equal to the number of observations. Integration of the relationships can change the contradictory patterns into more compatible ones which, in turn, can smoothly and quickly train the feed forward neural network with backpropagation algorithm (BP) to obtain the final relationship. A practical application is employed to show the superiority of the model. (C) 1999 Elsevier Science B.V. All rights reserved.