Apriori Summary
Frequent Itemsets
16| Size | Itemset | Support |
|---|---|---|
| 1 | bread | 77.78% |
| 1 | milk | 55.56% |
| 1 | beer | 44.44% |
| 1 | butter | 33.33% |
| 1 | diapers | 33.33% |
| 1 | chips | 22.22% |
| 2 | bread milk | 55.56% |
| 2 | bread butter | 33.33% |
| 2 | beer chips | 22.22% |
| 2 | butter milk | 22.22% |
| 2 | beer bread | 22.22% |
| 2 | bread diapers | 22.22% |
| 2 | diapers milk | 22.22% |
| 2 | beer diapers | 22.22% |
| 3 | bread butter milk | 22.22% |
| 3 | bread diapers milk | 22.22% |
Association Rules
12| Rule | Support | Confidence |
|---|---|---|
| {milk} → {bread} | 55.56% | 100.00% |
| {butter} → {bread} | 33.33% | 100.00% |
| {chips} → {beer} | 22.22% | 100.00% |
| {butter, milk} → {bread} | 22.22% | 100.00% |
| {diapers, milk} → {bread} | 22.22% | 100.00% |
| {bread, diapers} → {milk} | 22.22% | 100.00% |
| {bread} → {milk} | 55.56% | 71.43% |
| {butter} → {milk} | 22.22% | 66.67% |
| {diapers} → {bread} | 22.22% | 66.67% |
| {diapers} → {milk} | 22.22% | 66.67% |
| {diapers} → {beer} | 22.22% | 66.67% |
| {bread, butter} → {milk} | 22.22% | 66.67% |
Academic Output
Apriori finds frequent itemsets by starting from single items, keeping only those meeting a minimum support threshold, and then generating larger candidate itemsets by joining frequent itemsets of the previous size. Each candidate's support is computed by scanning transactions and counting how often the candidate is contained in a basket. The process repeats until no larger frequent itemsets can be found. Association rules are then derived from frequent itemsets by splitting an itemset into a left-hand side and right-hand side and calculating confidence as support(LHS ∪ RHS) divided by support(LHS).
This project demonstrates Association Rule Mining and Frequent Pattern Mining by extracting frequent itemsets and generating association rules using support and confidence metrics.