Predicting landslides spatial hazard with machine learning modelling.
This repository contains the analysis landslide hazard probability in Central Italy training, testing and deploying a Spatial Machine Learning model. The project aims to increase our understandment of landslide risk around use to guide environmental planning strategies.
- Analyze and prepare landslide records in Central Italy.
- Extracting several predictors for landslide occurrence.
- Investigate the relationship between predictors values of stable and unstable points.
- Train Gradient-Boosting model to predict locations with high-probability of future slope movements.
- Assess model accuracy and detection rate performance.
- Deploy the model to the entire area of iterest.
This project is licensed under the MIT License - see the LICENSE.md file for details.
Giandomenico Mastrantoni - giandomenico.mastrantoni@uniroma1.it