A Proposal for Improving Landslide Prediction with Machine Learning

• Renan Neves Rego UPE
• Thulio Aleixo Bezerra Brito UPE
• Keisy Lizandra Silva Vasconcelos UPE
• Fernando Pontual de Souza Leão Júnior UPE
• Jackson Raniel Florencio da Silva UPE

DOI: https://doi.org/10.5753/semish.2025.8501

Abstract

In Brazil, millions of people live in areas at risk of floods and landslides, especially in the Metropolitan Region of Recife. The increasing frequency of natural disasters, such as the heavy rains that caused deaths and damage in 2021 and 2022, highlights the need for effective alert systems. This study proposes the implementation of dynamic triggers in a hydrometeorological monitoring and landslide warning system, using machine learning to improve prediction accuracy and strengthen urban resilience. The results show that combining algorithms such as SVM with oversampling techniques like ADASYN achieved a recall of 0.91 for the landslide class, correctly identifying 10 out of 11 critical events, demonstrating the model’s potential to enhance event detection and mitigate risks through more precise monitoring. The main methodology used in this project is Design Science Research (DSR). In conclusion, the implementation of dynamic triggers, combined with the collection and standardization of relevant data, such as precipitation and landslide information, strengthens urban resilience and significantly contributes to the protection of lives and property. However, the study also emphasizes the need to expand the monitoring network, collect additional data, and improve the algorithms to ensure the system’s longterm accuracy and effectiveness.

References

Blagus, Rok, L. L. (2010). Class prediction for high-dimensional class-imbalanced data. BMC Bioinformatics, 11:523–523.

Breiman, L. (2001). Random forests. Machine Learning, 45(1):5–32.

Brunetti, M. T., Peruccacci, S., Rossi, M., Luciani, S., Valigi, D., and Guzzetti, F. (2010). Rainfall thresholds for the possible occurrence of landslides in italy. Natural Hazards and Earth System Sciences, 10(3):447–458.

Cortes, Corinna, V. V. (1995). Support-vector networks. Machine Learning, 20:273–297.

Costa, A., Cipriano, J., Rego, R., Oliveira, C., Orengo, J., Brito, C., Guedes, R., Silva, T., and Jr, F. (2023). DESENVOLVIMENTO DE UM SISTEMA DE ALERTA DE DESLIZAMENTOS DE TERRA EM ESCALA MUNICIPAL: MONITORAMENTO, GATILHOS E PREVISÃO METEOROLÓGICA.

de Assis Dias, M. C., Saito, S. M., dos Santos Alvalá, R. C., Stenner, C., Pinho, G., Nobre, C. A., de Souza Fonseca, M. R., Santos, C., Amadeu, P., Silva, D., et al. (2018). Estimation of exposed population to landslides and floods risk areas in brazil, on an intra-urban scale. International Journal of Disaster Risk Reduction, 31:449–459.

Friedman, J. H. (2001). Greedy function approximation: A gradient boosting machine. Annals of Statistics, 29(5):1189–1232.

Gregor, Shirley, H. A. R. (2013). Positioning and presenting design science research for maximum impact. MIS Quarterly, 37(2):337–355.

Han, Jiawei, Kamber, Micheline, Pei, and Jian (2012). Data Mining: Concepts and Techniques. Morgan Kaufmann, 3 edition.

He, H., Bai, Y., Garcia, E. A., and Li, S. (2008). Adasyn: Adaptive synthetic sampling approach for imbalanced learning. In 2008 IEEE international joint conference on neural networks (IEEE world congress on computational intelligence), pages 1322–1328. Ieee.

Hevner, A. R., March, S. T., Park, J., and Ram, S. (2004). Design science in information systems research. MIS quarterly, pages 75–105.

Hutter, F., Kotthoff, L., and Vanschoren, J. (2019). Automated machine learning: methods, systems, challenges. Springer Nature.

Jordan, M. I. and Mitchell, T. (2015). Machine learning: Trends, perspectives, and prospects. Science, 349:255–260.

Kanungo, D. and Sharma, S. (2014). Rainfall thresholds for prediction of shallow landslides around chamoli-joshimath region, garhwal himalayas, india. Landslides, 11(4):629–638.

Kohavi, R. (2001). A study of cross-validation and bootstrap for accuracy estimation and model selection. 14.

March, Salvatore, S. G. (1995). Design and natural science research on information technology. Decision Support Systems, 15:251–266.

Marchezini, V., o., editor (2017). Reduction of vulnerability to disasters: from knowledge to action. RiMa Editora São Carlos - SP.

Moraes, Katarina e Ferreira, C. (2023). Desastre das chuvas em pernambuco completa um ano com 134 mortes e nenhuma responsabilização. [link].

Nitesh, Chawla, V., Bowyer, K. W., Hall, L. O., and Kegelmeyer, W. P. (2002). Smote: Synthetic minority over-sampling technique. Journal of Artificial Intelligence Research, 16:321–357.

Peffers, K., Tuunanen, T., Rothenberger, M. A., and Chatterjee, S. (2008). A design science research methodology for information systems research. Journal of Management Information Systems, 24:45–77.