Predictive Modeling of Urban Air Pollution Using Machine Learning and Unmanned Aerial Vehicle Platforms
| dc.contributor.author | Özer, Muhammed Mirac | |
| dc.date.accessioned | 2025-03-17T12:22:44Z | |
| dc.date.available | 2025-03-17T12:22:44Z | |
| dc.date.issued | 2024 | |
| dc.department | Tarsus Üniversitesi | |
| dc.description.abstract | Air pollution poses a significant threat to human health on a global scale, with prolonged exposure to elevated ozone levels being particularly detrimental, leading to chronic respiratory conditions such as bronchitis, emphysema, and asthma. Beyond these health impacts, high ozone concentrations impair the photosynthetic capacity of plants, resulting in decreased agricultural productivity. Ozone is also a critical pollutant that negatively influences air quality in urban environments. In this regard, accurately predicting short-term air pollution levels is crucial for alerting the public to potential health risks and implementing effective pollution control measures. This study illustrates the successful prediction of hourly air pollutant concentrations in a specific region through the application of machine learning algorithms. Real-time pollutant and meteorological data—including air temperature, wind speed, relative humidity, and air pressure—collected via an unmanned aerial vehicle (UAV) were effectively utilized to develop the short-term forecasting model. Various machine learning regression algorithms, such as random forest, decision tree, support vector regression, k-nearest neighbors, and multilayer perceptron regression, formed the foundation of this predictive model. The analysis indicates that the random forest regression algorithm outperforms others in forecasting ozone levels in a particular area. Additionally, the data obtained from the UAV significantly enhances the accuracy and reliability of the short-term prediction model. The high-precision, instantaneous data provided by the UAV offers a considerable advantage in refining air pollution prediction models. These findings contribute to the development of robust strategies for mitigating air pollution and represent a significant advancement in adopting a more proactive and anticipatory approach to managing air quality issues. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024. | |
| dc.identifier.doi | 10.1007/978-3-031-69769-2_5 | |
| dc.identifier.endpage | 115 | |
| dc.identifier.issn | 1860-949X | |
| dc.identifier.scopus | 2-s2.0-85212130010 | |
| dc.identifier.scopusquality | Q3 | |
| dc.identifier.startpage | 79 | |
| dc.identifier.uri | https://doi.org/10.1007/978-3-031-69769-2_5 | |
| dc.identifier.uri | https://hdl.handle.net/20.500.13099/1335 | |
| dc.identifier.volume | 1171 | |
| dc.indekslendigikaynak | Scopus | |
| dc.institutionauthor | Özer, Muhammed Mirac | |
| dc.language.iso | en | |
| dc.publisher | Springer Science and Business Media Deutschland GmbH | |
| dc.relation.ispartof | Studies in Computational Intelligence | |
| dc.relation.publicationcategory | Kitap Bölümü - Uluslararası | |
| dc.rights | info:eu-repo/semantics/closedAccess | |
| dc.snmz | KA_Scopus_20250316 | |
| dc.subject | Air pollution forecasting | |
| dc.subject | Artificial intelligence | |
| dc.subject | Autonomous systems | |
| dc.subject | Data integration | |
| dc.subject | Dynamic data collection | |
| dc.title | Predictive Modeling of Urban Air Pollution Using Machine Learning and Unmanned Aerial Vehicle Platforms | |
| dc.type | Book Part |
