A GNN-based interpolation method for enhancing air pollution prediction based on Internet of Things (IoT) data

Key Points

• Enhanced prediction accuracy shows a reduction in forecasting error by 30%.
• This analysis employs a graph neural network model for air pollution estimation across multiple urban environments.
• The approach utilizes real-time IoT data for better pollution forecasting and analysis.
• Results highlight the potential for improved public health measures through accurate pollution monitoring.