Analysis of air quality pollution in Chengdu utilizing logistic regression for multi-class classification based on the distance metric between classes
Pollution Study · 2021
DOI: 10.54517/ps.v2i1.1976 ↗Kurz gefasst
Fachartikel in der Zeitschrift Pollution Study, 2021. Verfasst von Tingting Li und 3 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
The health of individuals is intimately linked to air quality, making it crucial to investigate the pollutants that influence it. A sequential Logistic multi-classification approach, which utilizes the distance between classes, was applied to analyze the air quality data of Chengdu spanning from May 2019 to April 2020. By leveraging the inter-class distance metric, the multi-class classification challenge was converted into a series of binary classification tasks. Using a sequential strategy, binary Logistic regression was then employed for each task. The accuracy rate post-stepwise regression was utilized to assess the impact of various pollutants on air quality. The findings indicate that PM2.5, PM10, NO2, and O3 are the four primary pollutants with the most significant collective influence on Chengdu's air quality. Consequently, the government should enhance the joint monitoring of these pollutants and develop targeted policies to mitigate their levels.
Zitieren
APA
Li, T., Lv, W., Zhou, J., & Zeng, S. (2021). Analysis of air quality pollution in Chengdu utilizing logistic regression for multi-class classification based on the distance metric between classes. Pollution Study. https://doi.org/10.54517/ps.v2i1.1976
Vancouver
Li T, Lv W, Zhou J, Zeng S. Analysis of air quality pollution in Chengdu utilizing logistic regression for multi-class classification based on the distance metric between classes. Pollution Study. 2021. doi:10.54517/ps.v2i1.1976
BibTeX
@article{Li2021,
author = {Li, Tingting and Lv, Wang and Zhou, Jiao and Zeng, Shijing},
title = {Analysis of air quality pollution in Chengdu utilizing logistic regression for multi-class classification based on the distance metric between classes},
journal = {Pollution Study},
year = {2021},
doi = {10.54517/ps.v2i1.1976}
}
EUVIMED