This is likely a simple question but I'm new to data mining techniques and am trying to compare two different predictive models. I've created a logistic regression and a classification tree and would like to look at the lift charts for both to compare their performance.
I understand how to construct the lift chart with a regression, I order the predictions of the validation set in descending order and compare it with the cumulative sums of the observed points.
However, for something like a classification tree where the output is a 1 or 0, how do I order the validation data points for the lift chart?