Rpart vs C50 - Regression tree vs Decision tree


I want to predict if a client is "Detractor" or not. So It is a binary output.
I have a dataset with 14 features. 2 of these features are integer type. My question is:

Should I use s c50 or rpart algorithm?
Which type o decision tree should I choose for the type of income features and binary output?
I know rpart is a regression tree model and c50 is a classification tree.

Thanks to all

rpart can be either classification or regression whereas C5.0 is only classification. You can use both and see what works; it's hard to know a priori which would be better for your data.

This topic was automatically closed 21 days after the last reply. New replies are no longer allowed.