Food and Beverage Menu Package Recommendation System Using Fp-Growth Algorithm

Abstract

In the era of digitalization, data utilization has emerged as a key to increasing company competitiveness in the current era of digital transformation, especially in the culinary industry. Recommendation systems are one way of analyzing data. Kedai Kopi Raja Palembang has so far only used transaction data as documentation without further analysis, making it difficult to determine sales products and customer purchasing patterns. This study aims to develop a menu package recommendation system using the Fp-Growth algorithm to analyze transaction data and provide more relevant recommendations. The development method used is the prototype method, with the provisions of the Knowledge Discovery in Database (KDD) stages which include data selection, preprocessing, transformation, application of the Fp-Growth algorithm, and evaluation using the Lift Ratio. The system was developed using Python and allows users to flexibly determine the minimum support and confidence values. The results of the study from the data used in the Raja Coffee Shop transaction history show that the system successfully identified 24 association rules from a minimum support of 1% and a Confidence of 70%. Then, an evaluation was carried out using the Lift Ratio which has a strong relationship between items and the 5 highest rules were taken from the Lift Ratio results for the recommendation menu. The implementation of the Fp-Growth algorithm in a transaction data-based recommendation system can enable business owners to make decisions by utilizing transaction data, to increase the efficiency of menu preparation, and become a model for other culinary businesses in optimizing marketing strategies.