Title: Utility-frequency analysis using customer market basket data in the retail sector

Authors: Gökçe Baysal Türkölmez; İpek Deveci Kocakoç

Addresses: Econometrics Department, Faculty of Economics and Administrative Sciences, Dokuz Eylul University, Turkey ' Econometrics Department, Faculty of Economics and Administrative Sciences, Dokuz Eylul University, Turkey

Abstract: In the retail sector, customers' shopping preferences are effective in determining the sales policies of stores. Due to the space problem, especially in small stores, it is not possible to store all the products as in big stores and to find places on shelves. The paper aims to make recommendations about product mix to small stores of a retail market chain in the Aegean region (Turkey). Six-month basket data of customers who have a membership card and who make purchases in regional stores of this market chain have been investigated by using frequent itemset mining and high utility itemset mining algorithms. Not only the data of frequency but also the utility data for items give useful information to determine the product mix of the small stores.

Keywords: data mining; customer market basket analysis; frequent itemset mining; high utility itemset mining.

DOI: 10.1504/IJBIS.2022.126064

International Journal of Business Information Systems, 2022 Vol.41 No.1, pp.69 - 81

Received: 21 Oct 2019
Accepted: 21 Dec 2019
Published online: 10 Oct 2022 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article