- Support: measures how often a set of items appears in transactions.
Milk and Bread appear together in 300 transactions → Support(Milk ∩ Bread) = 300 / 1000 = 30%
- Confidence : tells us how often B appears when A is present.
Confidence(Milk→Bread) = 0.30/0.40 = 0.75(75%)
If a customer buys milk, there is a 75% chance they also buy bread.
- Lift : measures how much more often A and B occur together compared to if they were independent.
Lift > 1 → A and B are positively correlated (A increases the likelihood of B and vice versa).
Lift = 1 → A and B are independent.
Lift < 1 → A and B are negatively correlated.
Customers who buy milk are 1.5× more likely to buy bread than a random customer.
Lift is symmetric: Lift(Milk → Bread) = Lift(Bread → Milk).
메모리를 더 효율적으로 사용하는 FP-V 알고리즘도 있음.
yozzum