Utilizing Sequential Pattern Mining and Complex Network Analysis for Enhanced Earthquake Prediction

Earthquakes are natural events caused by the movement of the earth's plates, often triggered by the energy release from hot liquid magma. Predicting earthquakes is crucial for raising public awareness and preparedness in seismically active areas. This study aims to predict earthquake activity by identifying patterns in seismic events using Sequential Pattern Mining (SPM). To enhance the prediction accuracy, Sequential Rule Mining (SRM) is applied to derive rules with confidence values from these patterns. The results show that using betweenness centrality as a weight increases the prediction accuracy to 83.940%, compared to 78.625% without weights. Using eigenvector centrality as a weight yields an accuracy of 83.605%. These findings highlight the potential of using centrality measures to improve earthquake prediction systems, offering valuable insights for disaster preparedness and risk mitigation.