ABSTRACT
This chapter addresses a few techniques that have been inspired by the metaphorical representation of a tree: From roots to branches and leaves. It discusses a clustering algorithm that organises groups of data in a hierarchical manner. The chapter also discusses decision trees, a tool that is widely used in decision analysis applications and operations research. Hierarchical clustering is an unsupervised learning task whose goal is to build a hierarchy of data groups. A decision tree is an algorithm described as non-parametrical because it does not require us to make any assumptions about parameters or distributions before starting our classification task. Decision trees are among the most well-known classification algorithms. A single tree provides a good shade, but the canopy of a group of trees is difficult to beat. If we grow a group of various decision trees as our base classifiers, and enable their growth to use a random effect we end up with a random forest.
