<![CDATA[The utilization of transaction data in footwear Micro, Small, and Medium Enterprises (MSMEs) is still often limited to administrative record-keeping, resulting in consumer purchasing patterns being underutilized as a basis for marketing strategies. This study aims to analyze consumer purchasing patterns in a footwear MSME in Binjai Village using the Frequent Pattern Growth (FP-Growth) algorithm and to develop data-driven product bundling recommendations based on actual transaction data. This study employs a quantitative data mining approach using Market Basket Analysis (MBA). The analyzed data consist of 3,549 actual transactions from Importshop Sofie Shoes during January–April 2026. The analysis stages include data cleaning, transformation into a binary matrix using one-hot encoding, frequent itemset extraction using FP-Growth, association rule generation, and evaluation based on support, confidence, and lift ratio. The results show that 3,522 transactions (99.24%) were single-product transactions, while 27 transactions (0.76%) were multiple-product transactions. This finding indicates that consumer purchasing behavior tends to be characterized by targeted shopping. Although multiple-product transactions were relatively limited, parameter adjustment enabled FP-Growth to identify positive association rules, indicated by a lift ratio greater than 1. These findings demonstrate that actual purchasing patterns can be utilized as a basis for developing product bundling and data-driven marketing strategies for footwear MSMEs. The study concludes that FP-Growth can identify relationships between products and support marketing decision-making in MSMEs.]]>
Copyrights © 2026
