Towards Smarter Cities: IoT and AI Integration for Effective Air Quality Monitoring and Forecasting

Background: The assessment of air pollution severity and its health effects heavily depends on Air Quality Index (AQI). The research implements Artificial Intelligence (AI) within the Air Quality Monitoring System (AQMS) to develop an economical air quality evaluation framework with extensive scalability. Methods: The implementation of a machine learning-based Random Forest model provides AQI value forecasts for comparative analysis with Multi-Layer Perceptron (MLP), Artificial Neural Networks (ANN), as well as AdaBoost on Kaggle data. The evaluation of model performance involved Mean Absolute Error (MAE) and Root Mean Square Error (RMSE) together with R<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> scores. Results: The Random Forest model delivered the best R<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> score at 98.08% together with a minimal Mean Square Error (MSE) of 1.97, which proved its exceptional predictive abilities. Conclusion: Furthermore, this work enables the use of AI and IoT capabilities for managing air quality, namely for ongoing monitoring and timely public health interventions.