الفهرس 
IntroductionOnly 14 pages are availabe for public view
 |  | from | 112 |  |  |
| | | from | 112 | | |
Abstract
This work presents how to predict earthquake magnitude using seismic earthquake data and some machine learning techniques in the region of southern California. Seven seismic indicators were mathematically and statistically calculated depending on pervious recorded seismic events in the earthquake catalogue of that region. These indicators are time taken during the occurrence of n seismic events (T), average magnitude of n events (M_mean), magnitude deficit that is the difference between the observed magnitude and expected one (ΔM), the curve slope for n events using inverse power law of Gutenberg Richter (b), mean square deviation for n events using inverse power law of Gutenberg Richter (η), the square root of the released energy during T time (〖DE〗^(1⁄2)) and average time between events (µ). Two hybrid machine learning models are applied here to predict the future magnitude during fifteen days. The first model is FPA-ELM which is the combination of Flower Pollination Algorithm (FPA) and Extreme Learning Machine (ELM).The second one is FPA-LS-SVM that is hybrid of FPA and Least Square Support Vector Machine (LS-SVM). The performance of the two models are compared and evaluated using four evaluation criteria Root Mean Square Error (RMSE), Mean Absolute Error (MAE), Symmetric Mean Absolute Percentage Error (SMAPE) and Percent Mean Relative Error (PMRE). The simulation results in most cases showed that FPA-LS-SVM model gave higher accurate magnitude prediction compared to the FPA-ELM model.