Customer segmentation using K-means clustering and the adaptive particle swarm optimization algorithm

The improvement of enterprise competitiveness depends on the ability to match segmented customers in a competitive market. In this study, we propose a customer segmentation method based on the improved K-means algorithm and the adaptive particle swarm optimization (PSO) algorithm. The current PSO al...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Applied soft computing Jg. 113; S. 107924
Hauptverfasser: Li, Yue, Chu, Xiaoquan, Tian, Dong, Feng, Jianying, Mu, Weisong
Format: Journal Article
Sprache:Englisch
Veröffentlicht: Elsevier B.V 01.12.2021
Schlagworte:
Adaptive parameter learning
Customer segmentation
K-means clustering algorithm
Mixed data
Particle swarm optimization algorithm
Customer segmentation
K-means clustering algorithm
Mixed data
Particle swarm optimization algorithm
Adaptive parameter learning
ISSN:1568-4946, 1872-9681
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:The improvement of enterprise competitiveness depends on the ability to match segmented customers in a competitive market. In this study, we propose a customer segmentation method based on the improved K-means algorithm and the adaptive particle swarm optimization (PSO) algorithm. The current PSO algorithm can easily fall into a local extremum; thus, adaptive learning PSO (ALPSO) is proposed to improve the optimization accuracy. On the basis of the analysis of population-based optimization, the inertia weight, learning factors, and the position update method are redesigned. To prevent the K-means clustering algorithm from depending on initial cluster centres, the ALPSO algorithm is used to optimize the K-means cluster centres (KM-ALPSO). Aimed at the issue of clustering the actual grape-customer consumption mixed dataset, factor analysis is used to extract numerical variables. We then propose a dissimilarity measurement method to cluster the mixed data. We compare ALPSO with several parameter update methods. We also conduct comparative experiments to compare KM-ALPSO on five UCI datasets. Finally, the improved KM-ALPSO (IKM-ALPSO) clustering algorithm is applied in customer segmentation. All results show that the three proposed methods outperform existing models. The experimental results also demonstrate the effectiveness and practicability of IKM-ALPSO for customer segmentation. •A new ALPSO algorithm is proposed to improve the optimization accuracy of PSO.•A new KM-ALPSO is proposed to avert the dependence of K-means on the initial centres.•A new IKM-ALPSO algorithm is proposed to address the issue of clustering mixed data.•All results show that the three proposed methods outperform existing models.•The results demonstrate the practicability of IKM-ALPSO for customer segmentation.
ISSN:1568-4946
1872-9681
DOI:10.1016/j.asoc.2021.107924