Pairwise basket association rules

Find directional item-pair rules with support, confidence and lift from transaction item lists.

The example value has been entered. Check it immediately or change the value.

Input is processed in the browser. Refreshing returns a sample value.

Result

Instructions and example

support(A→B)=count(A∩B)/N; confidence=count(A∩B)/count(A); lift=confidence/(count(B)/N)

For the default 4 transactions, bread→milk has support 0.5, confidence 2/3 and lift 8/9.

At most 100 transactions, 20 distinct items and 30 characters per name. Only single-item→single-item rules; duplicate or empty items within a transaction are rejected. Floating-point results display at most 12 significant digits.

Frequently asked questions

Why can A→B and B→A have different confidence?

Their joint count is the same, but the conditioning item count differs. Support agrees while confidence can differ.

Does high lift establish a purchase cause?

No. It describes association within transactions, not causation or guaranteed future purchases.

Calculation criteria and reference materials

Verification of calculation criteria: · Calculation·verification principles · Report errors