TEKNIK DATA MINING UNTUK PENENTUAN PAKET HEMAT SEMBAKO DAN KEBUTUHAN HARIAN DENGAN MENGGUNAKAN ALGORITMA FP-GROWTH (STUDI KASUS DI ULFAMART LUBUK ALUNG)

Abstract

In a business, it takes effort to maximize profits. Among them by doing promotions. Minimarket Ulfamart has not done innovation in the form of promotion of frugal packages to customers. The precision of promotion can be learned from the database of a retail company primarily shopping patterns on products commonly purchased simultaneously. Information about inaccurate customer spending patterns results in inaccurate and efficient promotional policies. One of the most common attempts to acquire and explore customer spending patterns is to use data mining known as Knowledge Discovery in Database (KDD). One of the data mining techniques is the Association Rule which is a procedure in Market Basket Analysis. Market basketball is defined as an itemset purchased simultaneously by the customer in a transaction. Market basket analysis is a powerful tool for cross-selling strategies. A pattern is determined by two parameters, namely support (value of support) and confidence (value of certainty). The Frequent Pattern Growth (FP-Growth) algorithm is used to help find some association rules from the database by applying a Tree Tree structure or called FP-Tree. Implementation using RapidMiner to help find accurate patterns to get a combination of items that can be used as a frugal package.