Tree Ensemble Predictor

Predicts patterns according to an aggregation of the predictions of the individual trees in a random forest model.

Random Forests is a registered trademark of Minitab, LLC and is used with Minitab’s permission.

Options

Change prediction column name

Select to customize the name of the column containing the prediction.

Prediction column name

Name of the output column containing the prediction.

Append overall prediction confidence

Adds the confidence of the predicted class; this is the maximum of all class confidence values, which can also be appended individually.

Append individual class probabilities

Adds one column per class containing its prediction confidence: the number of trees voting for that class divided by the total number of trees.

Suffix for probability columns

Suffix appended to the column names containing class probabilities.

Use soft voting

Switches from hard voting (the most votes win) to soft voting, which aggregates class probabilities from all trees. Requires the random forest model to store class distributions ("Save target distribution in tree nodes" in the learner); enabling this without stored distributions triggers a warning.

Views

This node has no views

Workflows

• _Predictor_FlowKNIME Hub
• _Predictor_FlowKNIME Hub
• 03_Active_Learning_Uncertainty_SamplingKNIME Hub
• 03_Learning_a_Tree_Ensemble_ModelKNIME Hub
• 04_TreeEnsembleKNIME Hub
• Show all 110 workflows

Links

• No links available

Developers