Application of Big Data-Driven Personalized Marketing Strategies in the Logistics Industry

This study presents the Multi-Layer Fusion Network (MLFN), a data-driven framework for enhancing precision marketing in logistics. It combines Temporal Encoder for data alignment, Cross-Domain Attention for cross-event associations, and Lightweight Auto-Encoder for compression and denoising. The framework incorporates an online gradient gating mechanism and Adaptive Recommendation Engine (AIE) to optimize real-time performance and improve customer conversion and repurchase rates. Results show significant reductions in inference latency and discount wastage, with a 12% increase in conversion and 15% in repurchase rates. MLFN outperforms baseline models in AUC, recall, and F1 scores. This framework offers an efficient solution for personalized marketing in logistics, with strong theoretical and practical value for digital transformation.