Integrating landslide magnitude in the susceptibility assessment of the City of Doboj, using machine learning and heuristic approach

In this work, landslide assessment of the Doboj City area was modeled by combining machine learning and heuristic tools. The part used to map Morphometric factor. i.e. probability landslides based on relation between magnitude events morphometric parameters: elevation, distance streams, slope, profile curvature, aspect. Random Forest Support Vector Machines algorithms were implemented in protocol, which included several strategies: balancing training/testing set size, algorithm optimization via cross-validation, cross-scaling. best performing factor ap created 50 m testing 25 dataset. for modeling Lithological Land Cover maps, expert-driven scoring their units, within 0-1 range values. final Susceptibility obtained multiplying all three maps resulting a high-performing model with AUC=0.97 acc=92%.