Version 1
: Received: 26 July 2024 / Approved: 27 July 2024 / Online: 30 July 2024 (09:03:24 CEST)
How to cite:
Panaite, F. A.; Rus, C.; Leba, M.; Ionica, A. C.; Windisch, M. Enhancing Air Quality Predictions on University Campuses: A Machine Learning Approach to PM2.5 Forecasting at the University of Petroșani. Preprints2024, 2024072290. https://doi.org/10.20944/preprints202407.2290.v1
Panaite, F. A.; Rus, C.; Leba, M.; Ionica, A. C.; Windisch, M. Enhancing Air Quality Predictions on University Campuses: A Machine Learning Approach to PM2.5 Forecasting at the University of Petroșani. Preprints 2024, 2024072290. https://doi.org/10.20944/preprints202407.2290.v1
Panaite, F. A.; Rus, C.; Leba, M.; Ionica, A. C.; Windisch, M. Enhancing Air Quality Predictions on University Campuses: A Machine Learning Approach to PM2.5 Forecasting at the University of Petroșani. Preprints2024, 2024072290. https://doi.org/10.20944/preprints202407.2290.v1
APA Style
Panaite, F. A., Rus, C., Leba, M., Ionica, A. C., & Windisch, M. (2024). Enhancing Air Quality Predictions on University Campuses: A Machine Learning Approach to PM2.5 Forecasting at the University of Petroșani. Preprints. https://doi.org/10.20944/preprints202407.2290.v1
Chicago/Turabian Style
Panaite, F. A., Andreea Cristina Ionica and Michael Windisch. 2024 "Enhancing Air Quality Predictions on University Campuses: A Machine Learning Approach to PM2.5 Forecasting at the University of Petroșani" Preprints. https://doi.org/10.20944/preprints202407.2290.v1
Abstract
This study focuses on predicting PM2.5 levels at the University of Petroșani by employing advanced machine learning techniques to analyze a dataset that encapsulates a wide array of air pollutants and meteorological factors. Utilizing data from IoT sensors and established environmental monitoring stations, the research leverages Random Forest, Gradient Boosting Machines, and Support Vector Regression models to forecast air quality, emphasizing the complex interplay between various pollutants. The models demonstrate varying degrees of accuracy, with the Random Forest model achieving the highest predictive power, indicated by an R2 score of 0.82764. Our findings highlight the significant impact of specific pollutants such as NO, NO2, and CO on PM2.5 levels, suggesting targeted mitigation strategies could enhance local air quality. Additionally, the study explores the role of temporal dynamics in pollution trends, employing time-series analysis to further refine the predictive accuracy. This research contributes to the field of environmental science by providing a nuanced understanding of air quality fluctuations in a university setting and offering a replicable model for similar environments seeking to reduce airborne pollutants and protect public health.
Copyright:
This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
