A Monitoring System to Prepare Machine Learning Data Sets for Earthquake Prediction Based on Seismic-Acoustic Signals

Estimating the location, time and magnitude of a possible earthquake has been the subject of many studies. Various methods have been tried using many input variables such as temperature changes, seismic movements, weather conditions etc. The relation between recorded seismic-acoustic data and occurring an anomalous seismic processes (ASP) has been proved in articles written by Aliev and et al. [1-4]. But it is difficult to predict the location, time and magnitude of the earthquake by using these data. In this study, it is aimed to prepare a data set/sets for prediction of an earthquake to be used in machine learning algorithms. An Earthquake-Well Signal Monitoring Software has been developed to construct these data sets. This study uses the on-line recordings of robust noise monitoring (RNM) signals of ASP from stations in Azerbaijan. An interface for analyzing the recordings and mapping them with previous earthquakes is designed