A "validation set lift chart"

a. None of the options are correct.

b. is similar to an "out of sample" test statistic used in classical business forecasting.

c. is rarely used to make an inference about the predictive capability of a k-Nearest-Neighbor model.

d. may not be used to judge the predictive power of a k-Nearest-Neighbor model.