Dan Wang, Feng Jiang, Zhanquan Wang
Ada-TransGNN is a novel air quality prediction model using adaptive graph convolutional networks and transformers to improve prediction accuracy and real-time updates.
Predicting air quality accurately is crucial for environmental monitoring and public health. The new model, Ada-TransGNN, combines advanced techniques from graph convolutional networks and transformers to better understand and predict air quality changes over time and across different locations. It uses a smart system to learn the best way to connect different air quality monitoring stations, making predictions more accurate. The model has been tested on standard datasets and a new dataset called Mete-air, showing that it performs better than current leading models for both short-term and long-term air quality forecasts.
