Random Forest Results
Click the OK button in the Random Forest Specifications dialog box to display the Random Forest Results dialog box. This dialog box contains various options for reviewing the final model. You can also specify to add more trees to the current model or select a specific number of trees for inclusion in the final results. In this case, all predicted values or classifications, as well as deployment code (for scoring or predicting new cases in predictive data mining applications), will be computed based on the requested number of trees.
The Random Forest Results dialog box contains four tabs: Quick, Classification, Prediction, and
. The Classification tab is available only if the current analysis is a Classification Analysis (selected on theRandom Forest
Startup Panel -Quick
tab).Random Forest Results
dialog box -Quick
tab to review a graph depicting the average squared error (regression analysis) or misclassification rate (classification analysis) for the training and test sample after the addition of a single tree. All results related to the computation of predicted values or classifications (those available on the Prediction or Classification tab, or when generating C/C++, SVB, or PMML code for deployment on the Report tab) will be generated consistent with this value, i.e. for the requested number of trees, and, hence.