A Short Survey of Human Mobility Prediction in Epidemic Modeling from Transformers to LLMs
Christian N. Mayemba, D'Jeff K. Nkashama, Jean Marie Tshimula, Maximilien V. Dialufuma, Jean Tshibangu Muabila, Mbuyi Mukendi Didier, Hugues Kanda, René Manassé Galekwa, Heber Dibwe Fita, Serge Mundele, Kalonji Kalala, Aristarque Ilunga, Lambert Mukendi Ntobo, Dominique Muteba, Aaron Aruna Abedi
TL;DR
This survey addresses the problem of forecasting human mobility during epidemic outbreaks to improve epidemic modeling and response planning. It surveys Transformer-based approaches and Large Language Models (LLMs) for mobility prediction, including architectures, datasets, and multimodal data integration, with notable examples like LP-BERT and LLM-Mob. The paper contributes a two-part taxonomy of mobility tasks (generation vs. prediction) and a comprehensive review of methods, architectures, and data sources, while discussing challenges such as data quality, generalization, and deployment in resource-constrained settings. The work underscores the potential of Transformer- and LLM-driven mobility modeling to inform public health decisions, resource allocation, and intervention strategies, while highlighting the need for responsible, context-aware deployment especially in LMICs.
Abstract
This paper provides a comprehensive survey of recent advancements in leveraging machine learning techniques, particularly Transformer models, for predicting human mobility patterns during epidemics. Understanding how people move during epidemics is essential for modeling the spread of diseases and devising effective response strategies. Forecasting population movement is crucial for informing epidemiological models and facilitating effective response planning in public health emergencies. Predicting mobility patterns can enable authorities to better anticipate the geographical and temporal spread of diseases, allocate resources more efficiently, and implement targeted interventions. We review a range of approaches utilizing both pretrained language models like BERT and Large Language Models (LLMs) tailored specifically for mobility prediction tasks. These models have demonstrated significant potential in capturing complex spatio-temporal dependencies and contextual patterns in textual data.