Abstract: Air quality fundamentally affects human wellbeing. Deterioration in air quality prompts an extensive variety of medical problems, mainly in young ones. The capacity to anticipate air quality allows the authority and also other related associations to take essential measures to protect the susceptible, while being prone to the air standards. Conventional ways of managing this problem have exceptionally limited success for the reason that an absence of access to such practices to adequate panel data. In this paper, we utilize a Support Vector Regression (SVR) model to predict the degrees of different contaminations and the Air Quality Index (AQI), utilizing chronicle pollution information which is made freely available by the Central Pollution Control Board and the US Embassy in New Delhi. Amidst the tried strategies, a Radial Basis Function (RBF) kernel generated the best outcomes with the SVR. In accordance with the analysis, utilizing the entire variety of accessible variables led to improved outcomes than utilizing characteristics chosen by principal component analysis. This model estimates levels of different pollutants, namely, nitrogen dioxide, sulfur dioxide, carbon monoxide, ground-level ozone, and particulate matter 2.5, as well as the Air Quality Index (AQI).
Keywords: Air Quality Index, Support Vector Regression, Radial Basis Function (RBF), Principal component analysis, Central Pollution Control Board
Works Cited:
Aishwarya Sajjan, M. Farida Begam, Ayush Dubey " Predicting Air Quality Index using most suitable ML model ", IJARCCE International Journal of Advanced Research in Computer and Communication Engineering, vol. 12, no. 10, pp. 106-114, 2023. Crossref https://doi.org/10.17148/IJARCCE.2023.121014
|
DOI:
10.17148/IJARCCE.2023.121015
