Market Based Analysis Sebagai Peningkatan Penjualan Produk Menggunakan Algoritma K-Medoids Dan Fp-Growth

Abstract

The COVID-19 pandemic has greatly affected the economy in Indonesia. The emphasis on the development of covid-19 through Enforcement of Restrictions on Community Activities has an impact on the Micro, Small and Medium Enterprises (MSME) sector, including product sales at the Distro Sextors. Therefore, a strategy is needed to increase sales, one of which is through data mining techniques with market based analysis. Data mining is one of the processes of Knowledge Discovery in Database (KDD) that uses scientific methods in its application. The data used in this study were 6531 sales transactions. The data mining algorithm used in this research is K-Medoids for grouping which aims to find the best cluster. The cluster experiments used are K-3, K-4, K-5, K-6 and K-7 then cluster testing is carried out using the Davies Bouldien Index (DBI) method and the results are cluster K-3 is the best cluster because it has a homogeneous value lowest. Then the association rule uses the FP-Growth algorithm and the minimum support value used is 6% to 7% while the minimum confidence value is 60%. The results of the association rule analysis are product recommendations that can be used as promotions to increase sales in Distro Sextors. The products are SE128, SE131, SE130, SE431, SE792, dan SE804.