Group-Aware Graph Neural Network for Nationwide City Air Quality Forecasting

TL;DR

This paper introduces GAGNN, a hierarchical graph neural network that models both spatial and latent dependencies among cities for accurate nationwide air quality forecasting, outperforming existing models.

Contribution

The paper proposes a novel group-aware GNN with a differentiable grouping network to capture latent city dependencies for nationwide air quality prediction.

Findings

GAGNN outperforms existing forecasting models on Chinese city data.

The model effectively captures dependencies between geographically distant cities.

Hierarchical modeling improves nationwide air quality forecasting accuracy.

Abstract

The problem of air pollution threatens public health. Air quality forecasting can provide the air quality index hours or even days later, which can help the public to prevent air pollution in advance. Previous works focus on citywide air quality forecasting and cannot solve nationwide city forecasting problem, whose difficulties lie in capturing the latent dependencies between geographically distant but highly correlated cities. In this paper, we propose the group-aware graph neural network (GAGNN), a hierarchical model for nationwide city air quality forecasting. The model constructs a city graph and a city group graph to model the spatial and latent dependencies between cities, respectively. GAGNN introduces differentiable grouping network to discover the latent dependencies among cities and generate city groups. Based on the generated city groups, a group correlation encoding module is introduced to learn the correlations between them, which can effectively capture the dependencies between city groups. After the graph construction, GAGNN implements message passing mechanism to model the dependencies between cities and city groups. The evaluation experiments on Chinese city air quality dataset indicate that our GAGNN outperforms existing forecasting models.