SISTEM PENDUKUNG KEPUTUSAN PENENTUAN RASA GELATO BEST SELLER MENGGUNAKAN METODE ALGORITMA APRIORI PADA CAFE LOUKOUMANNA

Penelitian Menggunakan Metode Algoritma Apriori

Abstract

The increasing amount of sales transaction data at Cafe Loukoumanna has not been optimally utilized as a basis for decision making. Sales data is generally only used as an archive without further analysis to identify customer purchasing patterns. This study aims to implement the Apriori Algorithm in a web-based Decision Support System to determine the best-selling gelato flavors based on historical transaction data. The research method adopted the CRISP-DM framework consisting of Data Understanding, Data Preparation, Modeling, Evaluation, and Deployment. The association rule mining process was conducted using RapidMiner with a minimum support value of 0.10 and a minimum confidence value of 0.80. The generated association rules were then implemented into a web-based application developed using Laravel and MySQL. The results indicate that the Apriori Algorithm successfully identifies purchasing patterns based on customer age, purchase day, cup size, scoop quantity, and flavor combinations. The developed system assists Cafe Loukoumanna in determining best-selling flavors and supports promotional and inventory management strategies based on transaction data.