3. Decision Trees, Random Forests, Support Vector Machines and Environmental Risk Analysis

In this chapter, the learning objectives are:

1. Training and benchmarking different types of support vector machines for classification and regression purposes,

2. Training, fine-tuning, and benchmarking decision trees and random forests for classification purposes,

3. Applying these classification algorithms to map wildfire risk in Italy using historical data of wildfire occurence.

Today’s tutorial:

1. Adapts Géron et al.’s Jupyter notebook exercises for chapters 5, 6, and 7 (License) of his book “Hands-On Machine Learning with Scikit-Learn, Keras, and TensorFlow, 2nd Edition”,

2. Adapts two articles on wildfire susceptibility mapping from Tonini et al. and Trucchia et al., and Python scripts from Giorgio Meschi.

If you are struggling with some of the exercises, do not hesitate to:

• Use a direct Internet search, or stackoverflow

• Ask your neighbor(s), the teacher, or the TA for help

• Debug your program, e.g. by following this tutorial

• Use assertions, e.g. by following this tutorial

If you’re done early, consider:

• Trying out the notebook’s bonus exercises

• Giving feedback on how to improve this notebook (typos, hints, exercises that may be improved/removed/added, etc.) by messaging the teacher and TA(s) on Moodle

• Working on your final project for this course.