Chunk #71 — The Methods — How Do Ensemble Methods Work? — Alternative Ensemble Methods

Source

An introduction to recursive partitioning: rationale, application, and characteristics of classification and regression trees, bagging, and random forests.

Embedded

yes

Text

Another very intuitive approach, that resides somewhere in between single classification trees and the ensemble methods we have covered so far, is TWIX (Potapov 2007; Potapov, Theus, and Urbanek 2006). Here the building of the tree ensemble starts in a single starting node but branches to a set of trees at each decision by means of splitting not only in the best cutpoint but also in reasonable extra cutpoints. A data driven approach for selecting extra cutpoints is suggested in Strobl and Augustin (2009).