The prediction of earthquakes in Western Anatolian region with artificial neural network method

A method that exactly predicts the earthquakes before their occurences and can generalize them have not been developed yet. However, earthquakes are tried to be predicted with numerous methods. Artificial neural networks, one of these methods, give appropriate outputs to different patterns by learning the relationship between the determined inputs and outputs. In this study, a feedforward back propagation artificial neural network that is connected to b_value - Gutenberg-Richter relationship and that bases on b_value used in earthquake predictions was developed. The artificial neural network was trained by using the earthquake data including four different regions which have intensive seismic activity in the Western Anatolian Region of Turkey. For this purpose, Gölhisar Çameli- Muğla Rodos region, Burdur Fault Zone, Büyük-Küçük Menderes Graben and Gediz Alaşehir Graben which have intensive seismic activity in the Western Anatolia are studied. After the training process, the earthquake data belonging to later dates of the same regions were used for testing and the success of the network was performed. When the prediction results of the developed network are examined, the prediction results that the network predicts that an earthquake will not occur are quite high in all regions. Furthermore, the results of the earthquake prediction that the network predicts that an earthquake will occur are different to some extent for the studied regions.