In data mining and association rule learning, lift is a measure of the
performance of a model (association rule) at predicting or classifying cases as
having an enhanced response (with respect to the population as a whole),
measured against a random-choice targeting model.
A measure of the usefulness of a rule. A greater value
means a better rule.
Microsoft is conducting an online survey to understand your opinion of the Technet Web site. If you choose to participate, the online survey will be presented to you when you leave the Technet Web site.