After the learning algorithm ranks all testing examples from most likely responders to least likely responders, we divide the ranked list in to 10 equal decile (could be more partitions), and see ho w the original resp onders distribute in the 10 deciles. If regularities are found, w e will see more responders in the top deciles than bottom deciles (Hughes, 1996). Table 2 is an example of the lift table. In this example, the testing set has a total of 100,000 examples (thus each decile has 10,000), with a total of 1,000 positive examples. Clearly, this lift table shows a distribution in which more responders are gathered in the top deciles than bottom deciles.