ANALISIS DATA PENJUALAN PADA TOKO BUKU SOCIAL AGENCY BARU AMBARUKMO MENGGUNAKAN METODE K MEANS CLUSTERING

Mohammad Ikhwanud Dawam, NIM.: 19106050050 (2023) ANALISIS DATA PENJUALAN PADA TOKO BUKU SOCIAL AGENCY BARU AMBARUKMO MENGGUNAKAN METODE K MEANS CLUSTERING. Skripsi thesis, UIN SUNAN KALIJAGA YOGYAKARTA.

Preview	Text (ANALISIS DATA PENJUALAN PADA TOKO BUKU SOCIAL AGENCY BARU AMBARUKMO MENGGUNAKAN METODE K MEANS CLUSTERING) 19106050050_BAB-I_IV-atau-V_DAFTAR-PUSTAKA.pdf - Published Version Download (2MB) | Preview
	Text (ANALISIS DATA PENJUALAN PADA TOKO BUKU SOCIAL AGENCY BARU AMBARUKMO MENGGUNAKAN METODE K MEANS CLUSTERING) 19106050050_BAB-II_sampai_SEBELUM-BAB-TERAKHIR.pdf - Published Version Restricted to Registered users only Download (4MB) | Request a copy

Abstract

SAB Ambarukmo Store is a prominent bookstore in Yogyakarta with a substantial customer base. This research was conducted with the aim of analyzing the survival rate of each product so that it is expected to increase the sales level in SAB Ambarukmo Store. Sales data was sampled for this study, and K Means Clustering was used to process the data. Several historical sales data from December 2022 to May 2023 were collected for this resear ch, if added up, there were 51948 total transactions that took place. After obtaining the data, it went through a preprocessing stage and resulted in 7947 data points. The data is then analyzed using the K Means Clustering method. The clustering process generates four outputs using different methods and numbers of clusters. The first output uses the K Means++ initialization method with three clusters, the second one uses the Forgy Initialization method with three clusters, and the third and fourth ones us e the K Means++ and Forgy Initialization methods, respectively, with two clusters each. The best outcome among the four generated outputs is achieved by using the K Means++ initialization method with two clusters, obtaining a silhouette score closest to 1 (

Share this knowledge with your friends :

Actions (login required)

View Item