Random Forests: some methodological insights

Robin Genuer (1, 2) , Jean-Michel Poggi (1, 2) , Christine Tuleau (3)

1 LM-Orsay - Laboratoire de Mathématiques d'Orsay
2 SELECT - Model selection in statistical learning
3 JAD - Laboratoire Jean Alexandre Dieudonné

This paper examines from an experimental perspective random forests, the increasingly used statistical method for classification and regression problems introduced by Leo Breiman in 2001. It first aims at confirming, known but sparse, advice for using random forests and at proposing some complementary remarks for both standard problems as well as high dimensional ones for which the number of variables hugely exceeds the sample size. But the main contribution of this paper is twofold: to provide some insights about the behavior of the variable importance index based on random forests and in addition, to propose to investigate two classical issues of variable selection. The first one is to find important variables for interpretation and the second one is more restrictive and try to design a good prediction model. The strategy involves a ranking of explanatory variables using the random forests score of importance and a stepwise ascending variable introduction strategy.

Mots clés: Random Forests; Regression; Classification; Variable Importance; Variable Selection

Domaines: Machine Learning [stat.ML]; Théorie [stat.TH]; Statistiques [math.ST]

Robin Genuer, Jean-Michel Poggi, Christine Tuleau. Random Forests: some methodological insights. [Research Report] RR-6729, INRIA. 2008. ⟨inria-00340725⟩