In this chapter, we will work with landslide data from Bangladesh and will create a landslide susceptibility map. We will use both R and QGIS to accomplish this. Using a logistic regression model, we will get the prediction probability of a landslide. The basics of model evaluation will also be discussed, and you will be introduced to the machine learning approach to modeling.
The following topics will be covered in this chapter:
- Preprocessing data in QGIS
- Further preprocessing data for model building using R
- Fitting logistic regression in R for predicting landslide and for susceptibility mapping
- Using Classification and Regression Tree (CART) and random forest for improving model accuracy
