<![CDATA[Increasing the efficiency of sales strategies and product stock management is a major requirement in the retail business, including in the sale of school uniforms. This research aims to identify consumer purchasing patterns through the application of the Market Basket Analysis method using two data mining algorithms, namely Apriori and Frequent Pattern Growth (FP-Growth). The approach used is CRISP-DM, consisting of six main stages, with a dataset of 365 sales transactions and minimum support parameters of 2% and confidence of 60%. The results showed that the Apriori algorithm generated association rules with an accuracy rate of 63.19%, average confidence of 75%, and support of 4.5%, while FP-Growth only achieved an accuracy of 2.92%. This finding shows that in the context of school uniform sales transaction data, Apriori is superior in exploring consumer purchasing patterns. The practical contribution of this research is the recommendation of product bundling and stock optimization strategies based on actual association patterns, which can be applied by educational retail businesses to improve business efficiency and effectiveness.]]>
Copyrights © 2025
