PENERAPAN ALGORITMA K-MEANS CLUSTERING UNTUK MENGANALISA TRANSAKSI PENJUALAN JASA CETAK PADA UNIT PRINT ON DEMAND (POD) PERCETAKAN GRAMEDIA

Abstract

Data mining is a method used to obtain valuable information contained in data banks. The information obtained can be used as input for determining the business strategy for the head of the company's agency. One of the most widely used data mining techniques is clustering using the K-Means method. Print on Demand (PoD) is one of the printing business units of PT. Gramedia which specifically provides printing services for various types of products, including: books, magazines, calendars, posters, promo materials such as product catalogs and brochures, as well as small-sized products such as business cards, tickets, coupons or vouchers and stickers. Currently every sales transaction data is stored in a SQL Server database, but until now the data processing is still done manually for reporting needs for company management. The purpose of this research is to perform K-Means Clustering analysis of the transaction data of sales of print services using the RapidMiner application to classify routine customer data based on the number of transactions made for each type of print product. The results of the application of the K-Means Clustering method resulted in 8 groups of customer data where the largest group consisted of 92.89% of the number of customers. The results of the analysis can be used by the company's management to determine various business strategies to increase the company's competitiveness.