<![CDATA[Fast-food restaurants generate large volumes of transaction data that can be utilized to understand customer purchasing behavior and support business decision-making. However, transaction data are often used only for operational reporting, limiting their potential for identifying product association patterns. This study aims to apply the Frequent Pattern Growth (FP-Growth) algorithm to discover frequent itemsets and association rules from burger restaurant transaction data and implement the results in a web-based application. The dataset used consists of 2,001 burger restaurant transactions collected from Kaggle, covering the period 2021–2023. The research process included data preprocessing, transaction transformation, FP-Tree construction, frequent itemset extraction, and association rule generation using a minimum support threshold of 2 transactions and a minimum confidence threshold of 60%. The results revealed that the most frequent items were Save Point Sundae (191 transactions), Health Potion Smoothie (181 transactions), and Cheat Code Cookies (164 transactions). Several association rules achieved a confidence value of 100%, indicating a strong co-occurrence relationship between products. Furthermore, the rules Avatar Avocado -> Cosmic Rings and Cosmic Rings -> Avatar Avocado obtained a lift ratio of 1.50, demonstrating a positive association between the two items. These findings indicate that FP-Growth is effective in identifying customer purchasing patterns and can support promotional strategies, product bundling, and inventory management through data-driven decision-making.]]>
Copyrights © 2026
