MACHINE LEARNING METHODS FOR EARTHQUAKE PREDICTION AND DAMAGE ASSESSMENT IN THE KYRGYZ REPUBLIC: A COMPARATIVE ANALYSIS OF APPROACHES AND A CONCEPTUAL ARCHITECTURE OF AN INTEGRATED SYSTEM
Keywords:
machine learning, earthquakes, prediction, damage assessment, Kyrgyz Republic, transformers, SegFormer, time series, seismologyAbstract
This article examines the application of deep learning methods to seismic data analysis and applied seismology problems. The main focus is on short-term prediction of earthquake parameters and assessment of damage consequences from satellite imagery. Data for the Kyrgyz Republic are used, and existing machine-learning methods for such tasks are analyzed.
Transformer models for time-series processing and image-segmentation architectures for identifying damaged areas are considered. Regional characteristics are taken into account, including complex geology, data imbalance, and a limited number of labeled satellite images, as well as shortcomings of existing approaches and solutions in this field.
Based on the analysis, a decision-support system concept is proposed to improve the effectiveness of emergency response.
References
1. Институт сейсмологии НАН КР. Сейсмичность Кыргызстана: каталог и мониторинг. — Бишкек: НАН КР, 2023.
2. Абдрахматов К.Е., Стром А.Л. Современные движения земной коры Тянь-Шаня и сейсмическая опасность // Геотектоника. — 2006. — № 3. — С. 44–57.
3. Mousavi S.M., Beroza G.C. Machine learning in earthquake seismology // Science. — 2022. — Vol. 377. — DOI: 10.1126/science.abm4470.
4. МЧС КР. Концепция снижения рисков стихийных бедствий КР на 2022–2030 гг. — Бишкек, 2022.
5. Münchmeyer J. et al. Which picker fits my data? // JGR Solid Earth. — 2022. — Vol. 127. — DOI: 10.1029/2021JB023499.
6. Merghadi A. et al. Machine learning methods for landslide susceptibility // Earth-Science Reviews. — 2020. — Vol. 208.
7. Breiman L. Random Forests // Machine Learning. — 2001. — Vol. 45. — P. 5–32.
8. Asim K.M. et al. Earthquake prediction model using support vector regressor // Natural Hazards. — 2017. — Vol. 85.
9. Wang Q. et al. Seismic event prediction using LSTM neural networks // Seismological Research Letters. — 2020. — Vol. 91.
10. Vaswani A. et al. Attention Is All You Need // NIPS. — 2017. — Vol. 30. — P. 5998–6008.
11. Woollam J. et al. SeisBench — A Toolbox for Machine Learning in Seismology // Seismological Research Letters. — 2022. — Vol. 93.
12. Woo G. et al. Graph Neural Networks for Seismic Activity Forecasting // Earth and Space Science. — 2023. — Vol. 10.
13. He K. et al. Mask R-CNN // Proceedings of ICCV. — 2017. — P. 2961–2969.
14. Xie E. et al. SegFormer: Simple and Efficient Design for Semantic Segmentation // NeurIPS. — 2021. — Vol. 34.
15. Tan M., Le Q. EfficientDet: Scalable and Efficient Object Detection // CVPR. — 2020.
16. Gupta R. et al. Creating xBD: A Dataset for Assessing Building Damage from Satellite Imagery // CVPR Workshops. — 2019.
17. Goodfellow I., Bengio Y., Courville A. Deep Learning. — MIT Press, 2016.
18. Hochreiter S., Schmidhuber J. Long Short-Term Memory // Neural Computation. — 1997. — Vol. 9.
19. Ridzwan N.S., Yusoff S.H. Machine learning for earthquake prediction: a systematic review (2017–2021) // Earth Science Informatics. — 2023. — Vol. 16.
20. USGS Earthquake Catalog. — URL: https://earthquake.usgs.gov/earthquakes/search/ (дата обращения: 01.03.2025).
21. Leema A., Balakrishnan P. et al. SeismoQuakeGNN: a hybrid framework for spatio-temporal earthquake prediction // Frontiers in Artificial Intelligence. — 2025. — Vol. 8. — DOI: 10.3389/frai.2025.1690476.
22. Norisugi R., Kaneko Y. et al. Machine learning-based prediction of meter-scale laboratory earthquakes // Nature Communications. — 2025.
23. Abebe W. et al. Earthquakes magnitude prediction using deep learning for the Horn of Africa // International Journal of Disaster Risk Reduction. — 2023.
24. Özdemir A. et al. Machine Learning Applications for Earthquake Magnitude Prediction in Western Turkey (1975–2024) // Applied Sciences. — 2025. — Vol. 15(20). — DOI: 10.3390/app152010909.
25. SEI/CMU. Improving Disaster Response with the xView2 Challenge. — Carnegie Mellon University SEI Report. — 2024. — URL: https://www.sei.cmu.edu/projects/xview-2-challenge/
26. Gupta A. et al. Forecasting Future Earthquakes with Machine Learning Models Based on Seismic Prediction Zoning // Applied Sciences. — 2025. — Vol. 15(24). — DOI: 10.3390/app152413116.
27. Чирягов, А. К. Методологические основы построения устойчивых моделей машинного обучения с использованием вероятностно-статистических методов / А. К. Чирягов, С. В. Корякин, К. Р. Карабакиров // Проблемы автоматики и управления. – 2026. – № 1(55). – С. 152-161. – EDN QLHKUW.
Downloads
Published
Issue
Section
Categories
License
Copyright (c) 2026 С.В. Корякин, К.И. Сабирова, М.Н. Ишенбиева
This work is licensed under a Creative Commons Attribution 4.0 International License.
