Predicting Landslide Susceptibility based on Naïve Bayes and Information Value Models in Shibam and AL-Taweelah District, Yemen

A study assessing landslide susceptibility in Shibam and Al-Taweelah districts, Yemen, reveals that the northern and northeastern regions are particularly vulnerable. Using machine learning (Naïve Bayes) and statistical models (Information Value), researchers analyzed 98 landslide events, achieving higher predictive accuracy with Naïve Bayes (AUC = 0.785) compared to Information Value (AUC = 0.674). This research provides critical insights for risk mitigation, construction planning, and disaster prevention in the UNESCO World Heritage-listed Old Walled City of Shibam.

Ghanem, O., Yi, H., Aljaadi, G., & Al-Qhfa, M. (2025). Predicting Landslide Susceptibility based on Naïve Bayes and Information Value Models in Shibam and AL-Taweelah District, Yemen. 2025 Third International Conference on Augmented Intelligence and Sustainable Systems (ICAISS), 1393–1397. https://doi.org/10.1109/icaiss61471.2025.11041763