Base Source files for the paper "Landslide susceptibility mapping using Machine Learning - A Danish case study"
Landslide susceptibility mapping using Machine Learning - A Danish case study
The preliminary mapping of landslides, conducted by GEUS, showed more than 3000 landslides in Denmark, indicating that landslides might be a bigger problem than previously acknowledged. Moreover, the changing climate is assumed to have an impact on landslide occurrences in the future. The aim of this project was to identify areas prone to landslides based on the chosen explanatory variables in an area of interest located around Vejle Fjord, Jutland. The ground truth data (locations of landslides) was obtained from GEUS's landslide database.
Three different machine learning algorithms - Random Forest, Support Vector Machine and Logistic Regression have been trained to classify the data samples as landslide or non-landslide, treating the Machine Learning task as a binary classification. The classification has been validated through the test data and through an external data set for an area located outside of the area of interest. While the performances varied slightly among the three models, the results show that the applied Machine Learning models have potential in landslide susceptibility mapping in Denmark. The mapping can potentially become a step on the way to planning for areas susceptible to landslides and for mitigating the potential risks associated with them.