datamining
Find the variables are related to churn rate, the churn rate is independent variable, and then using discriminant, logistic and cluster way to find out the cumulative lift. I want a perfect model on data set, which means the dataset classification results should be as high as possible, then we can get high cumulative lift.