Improvement of Fashion Product Sales Association Model in the Largest Store on Melgit Official Lazada with the Frequent Pattern Growth Algorithm
Year: 2025
Type: article
Abstract: In this digital era, it is increasingly easier for people to shop. E-commerce or Marketplace is a communication technology in the scope of business that seeks to maximize information from large transaction data to create relevant product recommendation models. The Melgit Official store on Lazada is one of the stores with the most sales of fashion products. To understand consumer buying patterns and improve sales strategies, data analysis is needed that can uncover associations between products that are frequently purchased together. One of the algorithms that can be used to find this association pattern is FP-Growth (Frequent Pattern Growth). The FP-Growth algorithm is effective in finding itemsets that frequently appear on purchase transactions to support business decision-making. which was chosen because of its efficiency in finding high-frequency patterns from large datasets without the need to perform repeated scans. By using the FP-Growth method, it is hoped that product purchase patterns can be found and can help in compiling product recommendations that are more in line with consumer preferences, as well as assisting stores in determining more effective sales strategies. This research was conducted with the aim of identifying the associations that sell the most fashion products in the Melgit Official store on the Lazada platform. This research method uses the FP-Growth approach, where the analyzed sales data results in product associations with certain support and confidence values. Support is measured as the proportion of total transactions that contain a particular item, while confidence indicates how often an item appears along with other items in a transaction. The results showed that the best support value in the FP-Growth algorithm was 0.1, which means that related products appeared in 10% of transactions. In addition, the confidence value obtained in the create association rules was 0.3, indicating that the association between the products in the formed rules was very strong and often occurred at the same time as the most product sales patterns, namely army rayon shirts and black rayon shirts. When buying an army rayon shirt, the probability of buying a black rayon shirt is 100%. From these results, it can be concluded that the FP-Growth algorithm provides significant results in determining the association pattern in fashion product sales data. This research is expected to help Melgit Official stores improve sales efficiency and customer satisfaction at Melgit Official stores, as well as contribute to the development of fashion product marketing strategies on e-commerce platforms.
Institution: STMIK IKMI Cirebon
Language: English
Subfield: Information Systems
Field: Computer Science
Domain: Physical Sciences
Open Access status: diamond
OpenAlex ID: W4407571266