Type your query or search by category
< All Topics

What is the difference between lift and leverage?

Both lift and leverage measure the relation between the probability of a given rule to occur (support(A→C)) and its expected probability if the items were independent (coverage(A)*coverage(C)) of each other. The lift calculates the ratio of these factors, i.e ( formula) and leverage calculates the difference, i.e (formula) The implications are that lift may find very strong associations for less frequent items, while leverage tends to prioritize items with higher frequencies/support in the dataset.

Table of Contents