R: Recipes for Analysis, Visualization and Machine Learning

Estimating the prediction errors of different classifiers

At the beginning of this chapter, we discussed why we use ensemble learning and how it can improve the prediction performance compared to using just a single classifier. We now validate whether the ensemble model performs better than a single decision tree by comparing the performance of each method. In order to compare the different classifiers, we can perform a 10-fold cross-validation on each classification method to estimate test errors using erroreset from the ipred package.

Getting ready

In this recipe, we will continue to use the telecom churn dataset as the input data source to estimate the prediction errors of the different classifiers.

How to do it...

Perform the following steps to ...