نوع مقاله : مقاله علمی -پژوهشی کاربردی
عنوان مقاله English
نویسنده English
This research has evaluated and modeled the stability of soil slopes in earthquake-prone areas using machine learning algorithms and intelligent methods. In this study, data collected from different regions of Iran, including 4295 soil slope samples with topographic, climatic, and geological characteristics, have been analyzed. The main goal of this research is to identify the most important parameters affecting slope stability and develop high-accuracy prediction models for landslide hazard simulation. Machine learning models such as Random Forest, XGBoost, and Gradient Boosting were evaluated to predict slope stability, and the results showed that the XGBoost model performed best with an accuracy of 0.920. Also, the Voting Classifier hybrid model, which is a combination of the three models, showed the best balance between accuracy and recall. This research can be used as an effective tool in assessing landslide hazards and assisting in regional planning and crisis management in earthquake-prone areas.
کلیدواژهها English