Deep Learning Approaches for Predictive Handover and Mobility Management in Ultra-Dense 5G Network Environments
Year: 2026
Type: conference-paper
Abstract: Matching the supply with the constantly changing demand is always a challenge for e-commerce firms since their demand depends on discounts, seasonality, and fast changes in consumer behaviour. This is because the fact that e-commerce data, in both respects, is relatively complicated and extensive makes it difficult to effectively use traditional methods of matching demand and supply. In order to do that, one should rely on advanced methods including those based on Long Short-Term Memory (LSTM) networks, Recurrent Neural Networks (RNNs), and Transformer models, as it is suggested in this article. At the moment, the cloud based deep learning models have proven themselves better than traditional techniques at forecasting the demand based on MAPE and RMSE scores for large e-commerce datasets. The study also reveals how these models could enhance order fulfilment efficiency, optimise inventory quantities and reduce stockouts. The issues and advantages in performance concerning the interpretation of model and real-time deployment are both brought to light in this study and a step toward using deep learning for demand-supply matching in e-commerce is provided. To augment further improvements, future research will study reinforcement learning integration and the use of multimodal data.
Language: English
Field: Decision Sciences
Domain: Social Sciences
Open Access status: hybrid
OpenAlex ID: W7203711032