PROJECT 21Advanced
Market Basket Analysis
Business scenario
Merchandising wants to know which products are bought together to design bundles and shelf placement.
Problem statement
Compute support, confidence and lift for product-family pairs and recommend bundles, distinguishing real associations from popularity effects.
Learning objectives
- Compute support, confidence and lift correctly
- Filter by minimum support
- Interpret lift > 1 and its limits
Instructions
Basket = order (completed). Item = subcategory (product family). Support(A) = baskets with A ÷ all baskets. Confidence(A→B) = support(A∩B) ÷ support(A). Lift = confidence(A→B) ÷ support(B).
Tasks
- Basket-item table at subcategory level.
- Pair counts with a self join (A < B).
- Support, confidence (both directions) and lift.
- Top rules with support ≥ 0.5%.
- Bundle recommendations and how to test them.
Deliverables
- SQL/Python
- Rules table
- Recommendation
Evaluation rubric
| Criterion | Weight | Excellent | Proficient | Developing | Insufficient |
|---|---|---|---|---|---|
| Data preparation & validation | 20% | Correct filters and grain; validation checks documented and totals reconciled. | Correct and complete with minor gaps in rigour or presentation. | Partly correct; important steps missing, unvalidated or unclear. | Missing, incorrect, or not supported by the data. |
| Metric correctness | 30% | Support, confidence and lift correct and directional. | Correct and complete with minor gaps in rigour or presentation. | Partly correct; important steps missing, unvalidated or unclear. | Missing, incorrect, or not supported by the data. |
| Interpretation | 15% | Lift vs popularity explained; test proposed. | Correct and complete with minor gaps in rigour or presentation. | Partly correct; important steps missing, unvalidated or unclear. | Missing, incorrect, or not supported by the data. |
| Interpretation & recommendations | 20% | Practical bundle recommendations. | Correct and complete with minor gaps in rigour or presentation. | Partly correct; important steps missing, unvalidated or unclear. | Missing, incorrect, or not supported by the data. |
| Presentation | 15% | Concise, well-labelled outputs a manager can act on without explanation. | Correct and complete with minor gaps in rigour or presentation. | Partly correct; important steps missing, unvalidated or unclear. | Missing, incorrect, or not supported by the data. |