Decision trees are good at handling interactions between variables.
How can you tell by looking at the rpart decision tree output what variables are correlated?
Decision trees are good at handling interactions between variables.
How can you tell by looking at the rpart decision tree output what variables are correlated?
You might want to look at random forests, which are ensembles of decision trees. I think they'd be more likely to give you correlations. See the randomForest and randomForestExplainer packages.
This topic was automatically closed 21 days after the last reply. New replies are no longer allowed.