<![CDATA[This study aimed to identify product purchasing patterns at Nofi Shop using the Frequent Pattern Growth (FP-Growth) algorithm and to implement the results as recommendations for product bundling, product arrangement, and promotional strategies. The research data consisted of 1,200 sales transactions recorded from September 2024 to May 2026. After preprocessing, 650 transactions that met the analysis criteria were obtained. The data were then transformed using One-Hot Encoding and processed using the FP-Growth algorithm with a minimum support of 4% and a minimum confidence of 60%. The analysis produced 35 association rules that met the criteria, with strong purchasing patterns such as Trash Bin and Broom, Mattress and Bed Sheet, and Chair and Table. The resulting association rules were subsequently implemented in a web-based system as recommendations for product bundling, product arrangement, and promotional strategies. Verification showed consistency among the manual calculations, Python implementation, and the developed system, while Black Box testing demonstrated that the main system functions operated as required. This study produced a system that could assist in identifying purchasing patterns and support data-driven decision-making at Nofi Shop.]]>
Copyrights © 2026
