Increasingly tight competition in the retail industry requires business owners to gain a deeper understanding of consumer behavior, particularly regarding which products tend to be purchased together. Sales transaction data collected daily actually holds hidden patterns that, when properly extracted, can support decision making such as shelf arrangement, bundling packages, and product recommendations. This study applies the Apriori algorithm, an association-rule-based data mining method, to analyze consumer purchasing patterns from sales transaction data. The research stages include data collection, preprocessing (cleaning and transforming data into binary tabular form), generating frequent itemsets using a minimum support threshold, and forming association rules using a minimum confidence threshold. Testing was carried out on sales transaction data with a minimum support of 20% and a minimum confidence of 60%. The results show that the Apriori algorithm is able to generate a number of strong association rules between products, which can subsequently be used as marketing strategy recommendations. The rule with the highest confidence value indicates that the purchase of one product is consistently followed by the purchase of another, suggesting a co-purchase tendency worth considering for cross-selling strategy and product placement.
Copyrights © 2025