Air Quality Forecasting using Near-Miss Approaches: A Predictive Model

One reporting metric for air quality is the air quality index (AQI). The purpose of the AQI is to alert the public, for a limited time, to the harmful impact that local air pollution has on health. Delhi's air pollution levels have significantly increased. The objective of this project is to determine how best to use prediction of AQI to assist temperature regulation. In order to guarantee that the most practical solution to the problem of air quality is found, this job combines a detailed investigation with the incorporation of Near Miss approaches. Three different techniques have been used in the proposed work: XGBoost regression (XGR), Knearest neighbour regression (KNNR), and decision tree regression (DTR) to determine the AQI of New Delhi. The investigation's findings demonstrate that, in terms of predicting New Delhi's air quality index, the XGBoost-NM model is clearly superior. It reaches the highest classification performance of 93.04% accuracy for AQI prediction by using the NM balanced approach. It is unequivocally shown that datasets treated with the Near Miss algorithm yielded results with greater accuracy.