This is the authors' abstract. We don't add key points for this paper.
This study presents a Hybrid Multi-Objective Framework that integrates Principal Component Analysis (PCA), Hierarchical Density-Based Spatial Clustering of Applications with Noise (HDBSCAN), Extreme Gradient Boosting (XGBoost), and the Non-dominated Sorting Genetic Algorithm II (NSGA-II) to enhance intelligent customer segmentation and Customer Lifetime Value (CLV) estimation. The framework jointly performs dimensionality reduction, behavioural segmentation, predictive modelling, and multi-objective optimization within a single analytical pipeline. PCA effectively condenses complex customer data into a compact representation that preserves the main behavioural patterns. HDBSCAN uncovers distinct clusters that correspond to high-value, retention-sensitive, and high-risk customer groups with clear behavioural differentiation. The XGBoost models demonstrate strong predictive capability in estimating both customer value and churn likelihood, enabling accurate forecasting of long-term profitability. NSGA-II optimization identifies a set of Pareto-efficient solutions that balance profitability improvement with churn reduction, providing a flexible decision space for strategic trade-offs. Overall, the integrated framework produces interpretable and actionable insights that support data-driven marketing strategies, personalized retention initiatives, and value-based decision-making.
Answers are grounded in this paper and cite their sources.
No takes yet. Share an insight, caveat, or question.
Elveny et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: