Point and pixel inclusive machine learning models for exploring gully erosion susceptibility

Sample point based spatial model derived from Machine Learning (ML) algorithms often generalizes the pattern of an event. The present study has tried to highlight how far it is acceptable and can replace with pixel modeling? presented a comparative view sample modeling gully erosion susceptibility upper Mayurakshi basin assess predictabilities. Random forest (RF), Support Vector (SVM), (ADB) have been applied for developing models in Python WEKA software environments respectively. show that 14–25% area located mainly parts unit very highly susceptible erosion. Based on accuracy performance level using under curve (AUC) Receiver operating curve, sensitivity, precision, F1 score, MCC, ensemble are superior modeling. point-based inferior agreement training testing data. So, could not be replaced models. RF found as best representative model. study, therefore, recommends this or similar purpose. Since figured out areas, would useful tool related planning processes.