Analisis Data Mining Menggunakan Algoritma K-Means Untuk Clustering Penjualan Studi Kasus Dapur Bu Ipung

Abstract

Ibu Ipung kitchen is a home industry that is engaged in making donuts located in Perum Cikarang Permai housing. Not only Bu Ipung's Kitchen, there are still quite a few other businessmen who are engaged in similar fields. This research is expected to be useful to help consumers see the products that are not selling well, selling well, and selling very well in the sales of Dapur Bu Ipung. The data collected from 2016-2021 is 570 sales transaction data. The purpose of this study is to group sales data into a cluster using the K-Means Clustering Algorithm Data Mining method. Sales data are grouped based on the similarity of the data so that data with the same characteristics will be in one cluster. So that it can be grouped using the K-Means algorithm into several criteria that are very good, sellable and less well-sold. The cluster that is formed after the K-Means Clustering process is carried out is divided into 3 Clusters Cluster 0 with 9 members with a percentage of 64.28% categorized as Less Selling, Cluster 1 number of members 1 with a percentage of 7.14% is categorized as very popular, and Cluster 2 with a percentage of 4 members with a percentage of 27.57% is categorized as popular, from the clustering process above, DBI validation (Davies Bouldin Index) can be obtained with a value of 0.451.

Keywords: : Data mining, Clustering, K-Means, Davies Bouldin Index