Enhanced Landslide Susceptibility Assessment in Angren, Uzbekistan using FR Feature Selection and CNN-GWO Hyperparameter Optimization

Key Points

• This research aims to enhance the accuracy of landslide susceptibility assessments in Angren, Uzbekistan.
• Utilized feature selection methods to identify key factors affecting landslide susceptibility.
• Applied convolutional neural networks (CNN) for predictive modeling.
• Employed grey wolf optimization (GWO) for hyperparameter tuning to improve model performance.
• Achieved higher accuracy rates in predicting landslide occurrences compared to traditional methods.
• Demonstrated the effectiveness of machine learning techniques in geological hazard assessments.