Generalised Random Forest Space Overview
This work provides a framework for researchers to more easily create new Random Forest variants, but it is incremental as it builds on existing generalizations without demonstrating new applications or data.
The authors tackled the problem of developing novel Random Forest methods by constructing a generalization space based on viewing Random Forests as nested ensembles of interchangeable modules, and they discussed module properties and existing generalizations.
Assuming a view of the Random Forest as a special case of a nested ensemble of interchangeable modules, we construct a generalisation space allowing one to easily develop novel methods based on this algorithm. We discuss the role and required properties of modules at each level, especially in context of some already proposed RF generalisations.