Fig. 5
From: A novel cost-sensitive framework for customer churn predictive modeling
Comparison of the distribution of the churn rate and the average unitary cost (C u =C 0/N) of each dataset with respect to the Total set. It is observed that the cost-sensitive sampling procedures have an average unitary cost much higher than the other sets, also showing why a simple under-sampling procedure of the dataset does not take into account the costs