Akademik Veri Yönetim Sistemi
7th International Conference on Applied Engineering and Natural Sciences ICAENS 2025, 15 - 16 Mayıs 2025, ss.1-6, (Tam Metin Bildiri)
Abstract – Avalanches are natural disasters
that can cause significant loss of life and property in mountainous regions.
This study investigates the identification of areas with avalanche risk using
machine learning-based methods. Within the scope of the analysis, topographical
features were initially considered, followed by the assessment of
meteorological data to determine risk levels. Various unsupervised learning
algorithms were tested on a labeled dataset containing existing avalanche
incidents in Kaçkar National Park, and the model performances were
comparatively analyzed. The results demonstrate that clustering and anomaly
detection-based methods are particularly effective in mapping regions at risk
of avalanches. This approach can contribute not only to winter sports
activities in our country but also provide critical insights for
decision-makers in pre-disaster prevention processes and land planning.