Web and Social Event Signals for AI-Driven Mobile Network Demand Forecasting

Background. Mobile network demand is increasingly volatile due to collective human activities such as concerts, sports events, and public gatherings. Traditional capacity planning methods, largely based on historical network indicators, struggle to anticipate these transient and localized demand surges. Recent advances in social sensing and artificial intelligence suggest that web and social signals can provide early indicators of real-world events, enabling proactive, event-aware network management in 5G and beyond. Methods. This paper presents a combined survey and empirical study on AI-driven event-aware demand forecasting for mobile networks. We first review methods for extracting event signals from web and social data and analyze prior evidence linking such signals to cellular traffic variations. We then introduce a forecasting-driven orchestration pipeline and evaluate it through a case study using the NetMob’23 dataset, which provides high-resolution, service-level mobile traffic traces from multiple urban areas. Several forecasting models—ranging from naïve baselines and linear regression to Random Forests and LSTM neural networks—are compared. We further investigate the impact of event-related features and introduce an asymmetric loss function designed to penalize traffic underestimation in proactive orchestration scenarios. Results. Results show that AI-based sequential models, particularly LSTM architectures, significantly outperform classical approaches in both prediction accuracy and operational effectiveness. Incorporating event-aware features reduces forecasting errors by up to 30% and yields substantial reductions in network overload under constrained capacity. The proposed asymmetric loss further improves robustness, nearly eliminating overload events at the cost of limited over-provisioning. Additional experiments demonstrate graceful degradation under noisy or unreliable event information.