Innovative Epidemic Forecasting with Graph Transformers

Created using ChatSlide

This project explores epidemic forecasting, emphasizing the importance of mobility-aware regional case predictions. It identifies gaps in existing literature on Spatio-Temporal Graph Neural Networks (STGNNs) and proposes a novel Graph Transformer solution. The methodology includes dataset preparation, model comparisons, and performance metrics. Experimental results from Spain and Brazil are analyzed, showcasing various forecasting metrics. The discussion highlights the implications of local...