AirCast: Improving Air Pollution Forecasting Through Multi-Variable Data Alignment

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Nedungadi, Vishal, Munir, Muhammad Akhtar, Rußwurm, Marc, Sarafian, Ron, Athanasiadis, Ioannis N., Rudich, Yinon, Khan, Fahad Shahbaz, Khan, Salman
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866929730120318976
author Nedungadi, Vishal
Munir, Muhammad Akhtar
Rußwurm, Marc
Sarafian, Ron
Athanasiadis, Ioannis N.
Rudich, Yinon
Khan, Fahad Shahbaz
Khan, Salman
author_facet Nedungadi, Vishal
Munir, Muhammad Akhtar
Rußwurm, Marc
Sarafian, Ron
Athanasiadis, Ioannis N.
Rudich, Yinon
Khan, Fahad Shahbaz
Khan, Salman
contents Air pollution remains a leading global health risk, exacerbated by rapid industrialization and urbanization, contributing significantly to morbidity and mortality rates. In this paper, we introduce AirCast, a novel multi-variable air pollution forecasting model, by combining weather and air quality variables. AirCast employs a multi-task head architecture that simultaneously forecasts atmospheric conditions and pollutant concentrations, improving its understanding of how weather patterns affect air quality. Predicting extreme pollution events is challenging due to their rare occurrence in historic data, resulting in a heavy-tailed distribution of pollution levels. To address this, we propose a novel Frequency-weighted Mean Absolute Error (fMAE) loss, adapted from the class-balanced loss for regression tasks. Informed from domain knowledge, we investigate the selection of key variables known to influence pollution levels. Additionally, we align existing weather and chemical datasets across spatial and temporal dimensions. AirCast's integrated approach, combining multi-task learning, frequency weighted loss and domain informed variable selection, enables more accurate pollution forecasts. Our source code and models are made public here (https://github.com/vishalned/AirCast.git)
format Preprint
id arxiv_https___arxiv_org_abs_2502_17919
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle AirCast: Improving Air Pollution Forecasting Through Multi-Variable Data Alignment
Nedungadi, Vishal
Munir, Muhammad Akhtar
Rußwurm, Marc
Sarafian, Ron
Athanasiadis, Ioannis N.
Rudich, Yinon
Khan, Fahad Shahbaz
Khan, Salman
Machine Learning
Computer Vision and Pattern Recognition
Air pollution remains a leading global health risk, exacerbated by rapid industrialization and urbanization, contributing significantly to morbidity and mortality rates. In this paper, we introduce AirCast, a novel multi-variable air pollution forecasting model, by combining weather and air quality variables. AirCast employs a multi-task head architecture that simultaneously forecasts atmospheric conditions and pollutant concentrations, improving its understanding of how weather patterns affect air quality. Predicting extreme pollution events is challenging due to their rare occurrence in historic data, resulting in a heavy-tailed distribution of pollution levels. To address this, we propose a novel Frequency-weighted Mean Absolute Error (fMAE) loss, adapted from the class-balanced loss for regression tasks. Informed from domain knowledge, we investigate the selection of key variables known to influence pollution levels. Additionally, we align existing weather and chemical datasets across spatial and temporal dimensions. AirCast's integrated approach, combining multi-task learning, frequency weighted loss and domain informed variable selection, enables more accurate pollution forecasts. Our source code and models are made public here (https://github.com/vishalned/AirCast.git)
title AirCast: Improving Air Pollution Forecasting Through Multi-Variable Data Alignment
topic Machine Learning
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2502.17919