Advancing PM 2.5 Forecasting: A Comparative Study of Ensemble Machine Learning Algorithms

Airborne fine particles, or PM 2.5, can quickly enter the respiratory system and lead to serious health issues like asthma, lung inflammation, eye and throat discomfort. When pollution is at its worst, such as in Delhi, PM 2.5 levels frequently above allowable limits. Predicting the PM 2.5 value ahead of time is crucial in order to provide environmental health experts and decision makers with relevant data and the ability to propose preventive measures. The goal of this study is to determine which machine learning algorithm such as XGBoost, AdaBoost, Random Forest, Gradient Boosting and HistogramGB is the most successful. In terms of computational efficiency during both the traning and prediction phases, the result validates XGBoost as the most effective algorithm in predicting PM 2.5 in Delhi. Additionally, the model has lower error rates.