The most important one of these methods is the so called “random forest” approach of Breiman (2001a): A random forest is a so called ensemble (or set) of classification or regression trees (CART, Breiman, Friedman, Olshen, and Stone 1984). Each tree in the ensemble is built based on the principle of recursive partitioning, where the feature space is recursively split into regions containing observations with similar response values. A detailed explanation of recursive partitioning is given in the next section.
