Store Segmentation in Retail Industry Using Clustering Algorithms


Unal A., Onal M., Kaya T., Özcan T.

4th International Conference on Intelligent and Fuzzy Systems (INFUS), Bornova, Turkey, 19 - 21 July 2022, vol.505, pp.409-416 identifier identifier

  • Publication Type: Conference Paper / Full Text
  • Volume: 505
  • Doi Number: 10.1007/978-3-031-09176-6_47
  • City: Bornova
  • Country: Turkey
  • Page Numbers: pp.409-416
  • Keywords: Store segmentation, Retailing, Clustering, K-means, Fuzzy C-means, Silhouette, Dunn
  • Istanbul Technical University Affiliated: Yes

Abstract

In today's digital age, the development of technology has made it easier for customers to reach everything. Store segmentation, which is one of the new methods, can be done in order to survive in the competitive environment due to the increase in retail companies. By doing this, they can gain an advantage by developing target marketing strategies specific to each segment instead of a whole marketing strategy. In this study, the data of 101 stores of a retail company were segmented according to 9 variables. These variables include the location of the stores, income levels, invoice numbers, inventory turnover, etc. has. Fuzzy C-means and K-means clustering algorithms were used for this study. Optimal cluster numbers were determined as 8 for Fuzzy C-means in terms of Dunn index and 7 for K-Means in terms of Silhouette index, which measure the effectiveness of clustering study.