Comparative Analysis of FPA-LS-SVM Model in Comparison with FPA-ELM, LS-SVM and ELM Model in Earthquake Prediction

This study presents a novel comparative analysis of several advanced machine learning algorithms for earthquake prediction. The focus is on the Flower Pollination Algorithm combined with FPA-LS-SVM and its performance in comparison with other models, including the Flower Pollination Algorithm combined with Extreme Learning Machine (FPA-ELM), Least Squares Support Vector Machine (LS-SVM), and Extreme Learning Machine (ELM). The aim is to evaluate which algorithm provides the best accuracy for predicting seismic events. The research involves implementing and evaluating these models to assess their predictive accuracy. Sample size calculations were conducted using G Power software, determining a sample size of 10 per group, with a pretest power of 80%, a significance threshold of 0.05%, and a confidence interval of 95%. The G Power values ranged between 0.59 and 0.9. The findings demonstrate that the FPA-LS-SVM model achieves an accuracy of 95.48%, surpassing the FPA-ELM, LS-SVM, and ELM models, which have accuracies of 93.18%. Statistical analysis reveals a significant difference (p < 0.05) between the FPA-LS-SVM model and the other models, with a significance value of 0.019. These results indicate that the FPA-LS-SVM model performs better in terms of accuracy compared to the FPA-ELM, LS-SVM, and ELM models.