Disaster Risk Perception and Personalized Learning in Natural Hazard Geography: AI- Adaptive Assessment Systems for Landslide and Flood Susceptibility Analysis.

The increasing frequency and severity of natural hazards such as landslides and floods have highlighted the growing importance of disaster risk perception and effective education within the field of geography. Understanding how individuals perceive risk is essential for improving preparedness and reducing vulnerability. However, traditional approaches to teaching natural hazards often depend on fixed content and uniform assessment methods, which do not adequately account for differences in students’ prior knowledge, spatial understanding, or learning pace. As a result, learners may struggle to fully grasp complex hazard processes and susceptibility patterns. This paper examines the role of AI-adaptive assessment systems in supporting personalized learning in natural hazard geography, with a specific focus on landslide and flood susceptibility analysis. Using a conceptual and model-based approach, the study draws on existing literature related to disaster risk perception, personalized learning, and artificial intelligence in education. A proposed framework demonstrates how adaptive assessments can adjust content difficulty, respond to individual learner profiles, and provide timely feedback through spatial and scenario-based tasks. The findings suggest that AI-driven adaptive assessment has the potential to enhance student engagement, improve spatial thinking skills, and promote a deeper understanding of hazard dynamics. These outcomes have important implications for geography education, as personalized and adaptive learning approaches can strengthen disaster awareness and contribute to more effective disaster preparedness and resilience-building efforts.