Innovative Earthquake Warning System for Enhancing Public Health Prevention

Abstract

An earthquake is a sudden ground shaking caused by tectonic activity, posing significant risks to human safety. Effective public health strategies focus on minimizing damage, providing timely warnings, and ensuring response plans to reduce health impacts during seismic events. The purpose of this research is to establish an innovative earthquake warning system for enhancing public health prevention. In this study, we propose a novel Mexican Axolotl Optimization-tuned Adjustable Support Vector Regression (MAO-ASVR) for accurately detecting the earthquake incidents. Our model integrates Internet of Things (IoT) technology to collect data from various sensors deployed across seismically active regions. The obtained signal data is pre-processed using Noise Filtering technique. In our proposed model, MAO optimization algorithm iteratively fine-tunes the ASVR architecture for enhancing the detection accuracy. The system activates an early warning alert within seconds of detecting seismic activity and sends notification, thereby minimizing potential damage. During the findings analysis phase, we evaluate our model's performance across various parameters. In addition, we also performed comparative analyses using diverse existing methodologies such as MAE with 0.397888, PMRE with 19.81432 errors, RMSE with 0.509074 errors. The findings demonstrate excellence and effectiveness of the suggested model.