Analisa Data Mining Menggunakan Frequent Pattern Growth pada Data Transaksi Penjualan PT Mora Telematika Indonesia untuk Rekomendasi Strategi Pemasaran Produk Internet

Harpa Erasmus Simanjuntak (Universitas Budi Luhur, Jakarta)
Windarto Windarto (Universitas Budi Luhur, Jakarta)

JURNAL MEDIA INFORMATIKA BUDIDARMA
Vol 4, No 4 (2020): Oktober 2020

Utilizing a lot of stored sales transaction data can provide useful knowledge in making policy and business strategy for PT Mora Telematika Indonesia. To realize the things can be applied with the Market Basket Analysis. Association Rule is a data mining technique which is a procedure in the Market Basket Analysis to find the knowledge of consumer purchase patterns. This pattern can be an input in making business policies and strategies. A pattern is determined by two parameters, which are support (supporting value) and confidence (value of certainty). In this study, the Market Basket Analysis used a Frequent Pattern Growth (FP-Growth) algorithm to find patterns by implementing TREE data structures or called FP-Tree. One of the patterns resulting from the analysis of data on sales transactions in the period of January 2018 to April 2018 is 7 Association rules with the highest lift ratio value is if there is an installation of OxygenHome 25-Super Double Then there will be installation OxygenHome 15-Super Double with elevator ratio 4.59%, support value of 3,125%, and confidence value 0.67%.