Neural-Powered Real-Time Air Quality Forecasting and Pollution Surveillance

This research introduces a novel methodology for air quality prediction that addresses thelimitations of traditional Air Quality Index (AQI) forecasting models by leveraging machine learning andenhanced secondary data modeling. The dataset utilized includes both forecast and actual measurements ofprimary pollutant concentrations and meteorological conditions, collected from monitoring stations inJinan, China, from July 23, 2020, to July 13, 2021. A comprehensive correlation analysis identified tenkey meteorological factors influencing pollutant concentrations, assessed through univariate andmultivariate techniques. Performance evaluation of various machine learning algorithms revealed theDecision Tree and Random Forest models achieving high accuracies of 99%. Additionally, the K-NearestNeighbors (KNN) classifier also demonstrated an accuracy of 99%, while Logistic Regression showed atraining accuracy of 72%. These findings affirm the reliability and efficacy of machine learning techniquesin enhancing air quality forecasting and underscore the importance of selecting appropriate algorithms foraccurate predictions.