Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Sports center customer segmentation: a case study

Juan Soto, Ramón Carmenaty, Miguel Lastra, Juan M. Fernández-Luna, José M. Benítez

TL;DR

This study tackles a sports-center customer segmentation problem characterized by a very large, heterogeneous dataset with substantial missing data. The authors implement a novel, region-wise unsupervised pipeline that starts with DBSCAN to estimate cluster structure and refines it with $k$-means, using a distance function whose feature weights are optimized by a genetic algorithm. Key innovations include data space partitioning to manage missingness, density-aware non-linear normalization, and an GA-driven weighting scheme guided by the Davies-Bouldin score, all implemented in Python with DEAP and scikit-learn. The approach yields 26 practically relevant customer segments from an initial 42, demonstrating improved clustering quality and actionable insights for targeted marketing in sports centers, while acknowledging the need for high-quality data and ongoing adaptation to evolving customer needs.

Abstract

Customer segmentation is a fundamental process to develop effective marketing strategies, personalize customer experience and boost their retention and loyalty. This problem has been widely addressed in the scientific literature, yet no definitive solution for every case is available. A specific case study characterized by several individualizing features is thoroughly analyzed and discussed in this paper. Because of the case properties a robust and innovative approach to both data handling and analytical processes is required. The study led to a sound proposal for customer segmentation. The highlights of the proposal include a convenient data partition to decompose the problem, an adaptive distance function definition and its optimization through genetic algorithms. These comprehensive data handling strategies not only enhance the dataset reliability for segmentation analysis but also support the operational efficiency and marketing strategies of sports centers, ultimately improving the customer experience.

Sports center customer segmentation: a case study

TL;DR

This study tackles a sports-center customer segmentation problem characterized by a very large, heterogeneous dataset with substantial missing data. The authors implement a novel, region-wise unsupervised pipeline that starts with DBSCAN to estimate cluster structure and refines it with -means, using a distance function whose feature weights are optimized by a genetic algorithm. Key innovations include data space partitioning to manage missingness, density-aware non-linear normalization, and an GA-driven weighting scheme guided by the Davies-Bouldin score, all implemented in Python with DEAP and scikit-learn. The approach yields 26 practically relevant customer segments from an initial 42, demonstrating improved clustering quality and actionable insights for targeted marketing in sports centers, while acknowledging the need for high-quality data and ongoing adaptation to evolving customer needs.

Abstract

Customer segmentation is a fundamental process to develop effective marketing strategies, personalize customer experience and boost their retention and loyalty. This problem has been widely addressed in the scientific literature, yet no definitive solution for every case is available. A specific case study characterized by several individualizing features is thoroughly analyzed and discussed in this paper. Because of the case properties a robust and innovative approach to both data handling and analytical processes is required. The study led to a sound proposal for customer segmentation. The highlights of the proposal include a convenient data partition to decompose the problem, an adaptive distance function definition and its optimization through genetic algorithms. These comprehensive data handling strategies not only enhance the dataset reliability for segmentation analysis but also support the operational efficiency and marketing strategies of sports centers, ultimately improving the customer experience.
Paper Structure (15 sections, 2 figures, 1 table, 1 algorithm)

This paper contains 15 sections, 2 figures, 1 table, 1 algorithm.

Table of Contents

  1. Introduction
  2. State of the art
  3. General Customer Analysis
  4. Customer Analysis in the Sports Activity Sector
  5. Sports centers: a case study
  6. Customer segmentation procedure
  7. Exploratory data analysis
  8. Data space partitioning
  9. Data preprocessing and transformation
  10. Determining the number of clusters
  11. Distance definition
  12. Genetic algorithms for distance definition optimization
  13. Clustering algorithm
  14. Experimental evaluation
  15. Conclusions

Figures (2)

  • Figure 1: Dataset division in regions
  • Figure 2: Graph describing the progress of fitness value over generations during the process of genetic optimization for region 5.