Title: Dynamic fuzzy clustering algorithm using cumulative cost curve with application to dynamic customer segmentation

Authors: M. Sivaguru; M. Punniyamoorthy

Addresses: Department of Management Studies, National Institute of Technology, Tiruchirappalli, India ' Department of Management Studies, National Institute of Technology, Tiruchirappalli, India

Abstract: Dynamic customer segmentation (DCS) is a useful tool for managers to update customer segments with new information continuously. One of the algorithms introduced in the literature for DCS is the dynamic fuzzy c-means (DFCM) clustering algorithm. We have proposed a cumulative cost curve method for the DFCM algorithm to overcome its shortcomings. Besides, we have also modified the existing structure strength expression used in the DFCM algorithm. Extensive experiments were conducted using a retail supermarket dataset with 11 new data updates to assess the performance of the proposed method in comparison with the DFCM and modified dynamic fuzzy c-means (MDFCM) algorithms. Finally, a DCS framework that could be used as a tool by managers is proposed to show the usefulness of the DFCM algorithm in the application domain, and it is demonstrated through a case study.

Keywords: dynamic fuzzy c-means; DFCM; dynamic customer segmentation; DCS; RFM analysis; cumulative cost curve; clustering error.

DOI: 10.1504/IJBIS.2024.138551

International Journal of Business Information Systems, 2024 Vol.46 No.1, pp.73 - 106

Received: 08 Feb 2020
Accepted: 05 Mar 2021
Published online: 12 May 2024 *

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