Analysis and statistical prediction of earthquake disaster impacts based on random forest algorithm
Yuxin Ji
University of Science and Technology Liaoning
DOI: https://doi.org/10.59429/esta.v13i2.14564
Keywords: random forest; earthquake disaster; disaster prediction; influencing factors; statistical analysis
Abstract
Earthquakes are severe natural disasters. Accurate analysis and prediction of disaster impacts are vital for disaster prevention and emergency response. Based on domestic and international earthquake data over the past two decades, this paper builds a prediction model with the random forest algorithm. Using descriptive statistics and correlation analysis, key influencing factors are identified. Model performance is verified via actual disaster data. The results show that the random forest algorithm achieves high accuracy and stability in predicting earthquake casualties and economic losses. Compared with traditional statistical methods, it has strong anti-interference ability and good generalization. The model can provide technical support for earthquake early warning, disaster assessment and emergency resource allocation.
References
[1] Li D P, Li M, Yin J F, et al. Research on Earthquake Disaster Scenario Construction Strategy from the Perspective of Social Perception[J]. China Emergency Rescue, 2024(6): 16-22. DOI:10.19384/j.cnki.cn11-5524/p.2024.06.012.
[2] Ai R W. Research on Comprehensive Statistical Analysis Technology and Method of Earthquake Emergency Survey Disaster Information[D]. Fuxin: Liaoning Technical University, 2014.
[3] Zhang P. Research on Post-earthquake Building Damage Assessment Model and Ground Motion Characteristic Analysis Method Based on Machine Learning[D]. Guangzhou: Guangzhou University, 2023. DOI:10.27040/d.cnki.ggzdu.2023.002187.
[4] Yu X. Basic Theoretical Research on Risk Assessment of Major Geotechnical Engineering[D]. Beijing: Institute of Engineering Mechanics, China Earthquake Administration, 2018.
[5] Li X A. Research on Earthquake Disaster Risk Assessment in Gansu Province[D]. Langfang: Institute of Disaster Prevention, 2021.
[6] Lin H, Du J L, Xu G, et al. Application of Random Forest Algorithm in Casing Deformation Prediction of Hydraulic Fracturing[J/OL]. Lithologic Reservoirs:1-9[2025-04-09]. http://kns.cnki.net/kcms/detail/62.1195.TE.20250211.1049.002.html.