Consumer purchase patterns based on market basket analysis using apriori algorithms

Abstract

Analyzing customer purchasing patterns can help minimarket expand marketing strategies by gaining insight into which items are often bought together by customers. Also, transaction data is a source of information available at the convenience store and one thing that can be used for business decision making. In this paper, we aim to use the Apriori algorithm method to obtain consumer purchasing patterns to analyze consumer purchasing patterns. This system uses a priori algorithm calculation method where the input data is consumer transaction data. Transaction data will be sorted and calculated by providing a minimum support value and a minimum confidence value. In this study, the authors conclude that the results of the analysis of information systems in determining consumer purchasing patterns can be as information to determine sales and in the application of Apriori algorithms can provide information in the form of a combination of consumer purchase patterns based on consumer transaction data with a minimum support value above 10% and the minimum confidence value is above 65%.

Toward Enhanced Customer Transaction Insights: An Apriori Algorithm-based Analysis of Sales Patterns at University Industrial Corporation

“…In marketing, its utility extended to the dissection of purchase transaction-based product promotions [8], recommending ecommerce links [2], and characterizing e-customer behavior to discern purchase patterns among two customer groups [3]. Noteworthy studies have included analysis of purchased products [11], examination of consumer purchasing patterns [12], encompassing market basket analysis [13], and online shopping through RFMDR model [14]. Furthermore, it found practical application in minimum spanning tree-based shopping [15], and Internet marketing strategies in the cosmetics sales in Taiwan [16].…”

Algoritma Apriori Dalam Identifikasi Pola Pembelian Konsumen Pada Produk Minuman

“…Maka, dilakukanlah pengujian menggunakan lift ratio untuk memperkuat atau tidaknya aturan asosiasi tersebut. Jika nilai lift ratio bernilai 1 atau lebih, maka aturan asosiasi tersebut kuat dan terdapat keterikatan yang kuat antar satu sama lain [14]. Lift ratio diperoleh membagi nilai confidence dengan nilai expected confidence.…”

Retail Customer and Market Proclivity Assessment using Historical data and Social Media Analytics

“…Basket analysis (Efrat, A. R., & Gernowo, R., 2020;Kabasakal, İ., 2020) is a longstanding task in the area of machine learning. Association rule mining (Agrawal, R. et al, 1993;Videla Cavieres, I. F., 2014;Suharjo, R. A., & Wibowo, A., 2020) has been the dominating state-of-the-art in terms of retail analysis for a couple of years.…”