Penerapan Data Mining Pada Penjualan Produk Digital Konter Leppangeng Cell Menggunakan Metode K-Means Clustering

Abstract

The Leppangeng Cell counter is a counter that sells digital products in the form of Credit, Data Packages, PLN Electricity Tokens, Electricity Payments, BPJS Payments, PDAM Payments. Spidy or Indihome Payments, Home Phone Payments, Postpaid Payments, Game Voucher Sales, Cable TV Payments, Credit Installment Payments, Train Ticket Payments, PGN/Gas Payments, Shope Payments and Cash Transfer Services and Top Up. However, from the counter sales, of course, not all of them sell well, there are also products that don't sell. Sales data, purchases and unexpected expenses at the Leppangeng Cell Counter are not well structured, so the data only serves as a store archive and cannot be used to develop marketing strategies. Therefore, it is necessary to apply data mining using K-Means at the Leppangeng Cell Counter. The K-Means method can be applied to the Leppangeng Cell Counter to determine sales of digital products that are selling well and not selling well. The application of the K-Means method at the Leppangeng Cell Counter is by grouping digital product stock data. Then select 3 random grub as initial centroid. After the data for each group does not change, the results of data processing are obtained in the form of 114 products that sell, 5 underperforming, and 14 not sold. Then, the application of the K-Means method to Rapidminer is carried out by entering the initial product stock, the sold stock and the final stock to be changed and form K-Means. After that, Rapidminer will produce which products have high and low demand