Intelligent Earthquake Prediction System Based On Neural Network

Predicting earthquakes is an important issue in the<br> study of geography. Accurate prediction of earthquakes can help<br> people to take effective measures to minimize the loss of personal<br> and economic damage, such as large casualties, destruction of<br> buildings and broken of traffic, occurred within a few seconds.<br> United States Geological Survey (USGS) science organization<br> provides reliable scientific information about Earthquake Existed<br> throughout history &amp; the Preliminary database from the National<br> Center Earthquake Information (NEIC) show some useful factors to<br> predict an earthquake in a seismic area like Aleutian Arc in the U.S.<br> state of Alaska. The main advantage of this prediction method that it<br> does not require any assumption, it makes prediction according to the<br> future evolution of the object's time series. The article compares<br> between simulation data result from trained BP and RBF neural<br> network versus actual output result from the system calculations.<br> Therefore, this article focuses on analysis of data relating to real<br> earthquakes. Evaluation results show better accuracy and higher<br> speed by using radial basis functions (RBF) neural network.