Search code examples
raprioriarules

Arules in R: A very high measure for lift

Running Apriori with the Arules package, generates rules with very high lift.

For example:

A -> B support=0.0023 confidence=0.6832 lift=28.02

(min_support=0.002 and min_conf=0.2)

In some rules, the lift is as high as 250! I have seen people discuss a lift greater than 1 (most of them <5), but I'm wondering how I can interpret these rules with very high lift values.

Solution

  • The lift measure stands for how likely item Y is purchased when item X is purchased, while controlling for how popular item Y is:

    support(X,Y)/(support(X)*support(Y)).

    Generally, a lift value greater than 1 means that item Y is likely to be bought if item X is bought, while a value less than 1 means that item Y is unlikely to be bought if item X is bought. If you had a huge dataset, that would be interesting, however, with your small dataset and support, this is just nothing. Check here for a nice illustrative answer: