T4.2.1 - Short-Term Pollution Prediction Using Personal Environmental Monitoring and Machine Learning
- Event
- EUROSENSORS XXXVII
2025-09-07 - 2025-09-10
Wroclaw - Band
- Lectures
- Chapter
- T4.2 - AI for Sensors
- Author(s)
- F. Pan, J. A. Covington - University of Warwick United Kingdom)
- Pages
- 110 - 111
- DOI
- 10.5162/EUROSENSORS2025/T4.2.1
- ISBN
- 978-3-910600-07-2
- Price
- free
Abstract
Due to increasing global concerns around air pollution, research on portable environmental monitoring has become an expanding area of research. This study used an in-house developed personal environ-ment monitor, coupled to machine learning, to identify scene switch recognition and short-term pollution prediction. Here, decision trees excelled in environmental identification (97% accuracy), while XGBoost performed best in pollution prediction (R² = 0.93). This study provides an effective scheme for short-term pollution warnings, which in the future could be used to warn users of potential harm.