Dynamic customer segmentation: a case study using the modified dynamic fuzzy c-means clustering algorithm

Dynamic customer segmentation (DCS) is a useful tool for managers to adjust their marketing strategies from time to time. However, no study in the literature has attempted to develop a DCS framework until now. To fill the research gap, a DCS framework is proposed. To improve the effectiveness of the...

Full description

Saved in:
Bibliographic Details
Published in:Granular computing (Internet) Vol. 8; no. 2; pp. 345 - 360
Main Author: Sivaguru, M.
Format: Journal Article
Language:English
Published: Cham Springer International Publishing 01.03.2023
Springer Nature B.V
Subjects:
Algorithms
Artificial Intelligence
Clustering
Computational Intelligence
Customers
Engineering
Managers
Original Paper
Segmentation
Statistical tests
Supermarkets
Dynamic customer segmentation
means
Dynamic fuzzy
RFM analysis
Retail supermarket
ISSN:2364-4966, 2364-4974
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Dynamic customer segmentation (DCS) is a useful tool for managers to adjust their marketing strategies from time to time. However, no study in the literature has attempted to develop a DCS framework until now. To fill the research gap, a DCS framework is proposed. To improve the effectiveness of the proposed framework, the existing dynamic fuzzy c -means clustering (dFCM) algorithm is modified owing to certain limitations found in it. Extensive experiments were conducted using the retail supermarket dataset to assess the performance of the modified dFCM (MdFCM) algorithm. Experimental results prove that the MdFCM algorithm performs better than the existing algorithm. The experimental results are validated by fuzzy clustering evaluation measures such as Xie–Beni index (XBI), partition coefficient (PC), modified partition coefficient (MPC), partition entropy (PE), and fuzzy silhouette index (FSI). A statistical significance test, MANOVA Pillai's statistics, is carried out to prove that clusters obtained from the MdFCM algorithm are significant. Finally, a case study on a retail supermarket has been conducted using the proposed DCS framework. The study has shown that the proposed DCS framework extracts useful information for managers to support strategic decision-making.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:2364-4966
2364-4974
DOI:10.1007/s41066-022-00335-0