intelligent classification of stable and unstable slope conditions based on landslide movement

One of the most critical problems in the study of geohazards is the displacement brought on by landslides. this research aims to investigate stable and unstable conditions for this important issue using new techniques. there are several effective parameters on landslide movement that need to be thoroughly investigated/observed, making the process of determining the movement of landslides a difficult one. in this research, different machine learning-based approaches were used to analyze and manage this problem. a set of data was compiled for this investigation including groundwater level, prior rainfall, infiltration coefficient, shear strength, and monitored slope gradient are all influential in landslide movement. three models of tree, adaboost and artificial neural network (ann) were developed for classification into two categories, stable and unstable. the results showed well that two adaboost and tree models can provide significant performance for determining stable and unstable conditions. for the test data, the adaboost model with an accuracy of 0.857 has the highest accuracy, followed by the tree model with an accuracy of 0.786. finally, in this research, unstable data using machine learning was used to evaluate and predict the amount of slope movement. this system is well suited for its high flexibility and high-accuracy assessment for conditions with more movement.