Skip to main content
. Author manuscript; available in PMC: 2010 Jul 6.
Published in final edited form as: J Proteome Res. 2009 Jul;8(7):3558–3567. doi: 10.1021/pr900253y

Figure 1.

Figure 1

The Random Forest classifier is an ensemble of decision trees where the single trees are constructed from bootstrap samples. On the left and in the center, two trees of the forest are shown in detail: At each node, the feature which allows for the best class separation is chosen (with respect to the subset of features selected for that node). The corresponding partitioning of the feature space is shown below with the decision boundary plotted in purple. On the right, the decision boundary of the Random Forest is displayed. It is based on the majority votes of the individual trees.