Article: Real-time customer segmentation using big data and cluster analysis in enterprise marketing strategies Journal: International Journal of Information and Communication Technology (IJICT) 2026 Vol.27 No.51 pp.1 - 16 Abstract: Addressing the issue of traditional customer segmentation relying on static data and struggling to respond to behavioural changes in real-time, a real-time customer segmentation framework based on big data analysis and clustering analysis is proposed. The data comes from e-commerce websites and includes user activities, transactions, and demographic information. Preprocessing involves data cleaning, normalisation, and TF-IDF feature extraction. The key features include transaction frequency, interest in product categories, and page dwell time. The proposed model is an adaptive k-nearest neighbour (k-NN) logistic regression based on clonal selection (CS-AK-LR), integrating adaptive K-means clustering (AK) and logistic regression (LR) for customer clustering and value classification prediction. The clonal selection algorithm (CS) optimises the hyperparameters of AK and LR. The segmentation detection rate of this method reaches 96.21%, and the error rate is reduced by 1.03% compared to existing methods. Combining big data with real-time clustering analysis can effectively enhance the speed and accuracy of marketing responses. Inderscience Publishers - linking academia, business and industry through research

Title: Real-time customer segmentation using big data and cluster analysis in enterprise marketing strategies

Authors: Zhengyang Fang; Neng Wan

Addresses: College of Management, Ningbo University of Finance and Economics, Ningbo, Zhejiang, China ' Hangzhou Neuroiz Co., Ltd., Hangzhou, Zhejiang, China

Abstract: Addressing the issue of traditional customer segmentation relying on static data and struggling to respond to behavioural changes in real-time, a real-time customer segmentation framework based on big data analysis and clustering analysis is proposed. The data comes from e-commerce websites and includes user activities, transactions, and demographic information. Preprocessing involves data cleaning, normalisation, and TF-IDF feature extraction. The key features include transaction frequency, interest in product categories, and page dwell time. The proposed model is an adaptive k-nearest neighbour (k-NN) logistic regression based on clonal selection (CS-AK-LR), integrating adaptive K-means clustering (AK) and logistic regression (LR) for customer clustering and value classification prediction. The clonal selection algorithm (CS) optimises the hyperparameters of AK and LR. The segmentation detection rate of this method reaches 96.21%, and the error rate is reduced by 1.03% compared to existing methods. Combining big data with real-time clustering analysis can effectively enhance the speed and accuracy of marketing responses.

Keywords: consumer segmentation; clonal selection-based adaptive K-logistic regression; CS-AK-LR; marketing strategy; big data; cluster analysis.

DOI: 10.1504/IJICT.2026.153705

International Journal of Information and Communication Technology, 2026 Vol.27 No.51, pp.1 - 16

Received: 21 May 2025
Accepted: 02 Jul 2025
Published online: 21 May 2026 *

Title: Real-time customer segmentation using big data and cluster analysis in enterprise marketing strategies

Keep up-to-date