Applying Machine Learning Techniques for Weather and Air Pollution Forecasting with Geographically Distributed Data
The weather conditions and air pollution data can be closely related as they both represent atmosphere as an object from physical world. The usage of combined datasets is expected to improve the forecasting accuracy. However, this still needs to be verified. This article studies the 12-hour ahead forecasting of weather and air pollution parameters for Odesa city, the cross-influence of these parameters and potential accuracy gain when dataset is augmented with parameters from nearby geographies.
Authors
- А.Yu. Doroshenko (ORCID: https://orcid.org/0000-0002-8435-1451)
- Vitalii Shpyg (ORCID: https://orcid.org/0000-0003-1055-7120)
- Ivan Dimov (ORCID: https://orcid.org/0000-0003-4198-5830)
- Róbert Horváth (ORCID: https://orcid.org/0000-0001-8617-2302)
- Dmitry Zhora (ORCID: https://orcid.org/0009-0006-6073-7751)
- Y.O. Haidukevych
Institutions
- Budapest University of Technology and Economics (HU)
- Ukrainian Hydrometeorological Institute (UA)
- Institute of Mathematics and Informatics (BG)
- Institute of Software Systems (UA)
- Institute of Information and Communication Technologies (BG)
Publication Details
- Published
- 2025-08-07
- DOI
- https://doi.org/10.1109/acdsa65407.2025.11166239
- Primary Topic
- Air Quality Monitoring and Forecasting
- Type
- article
- Field-Weighted Citation Impact
- 0.00