Nature inspired artificial intelligence (AI) techniques are gaining traction in air quality modeling. A mathematical simulation of how air pollutants distribute and react in the atmosphere to impact ambient air quality is known as air quality modeling. Modeling aids in estimating the correlation between pollution levels and their impacts on air quality. Outdoor ambient air quality is important because it affects indoor air quality. Unprecedented growth of urban agglomeration, population blast, and unrestricted use of natural resources as well as nonrenewable fuels eventually results in higher pollution and driving the world away from the sustainable development goals. Rising air pollution levels around the globe is an alarming situation for all lifeforms. It also affects the nonliving things such as structures, metals, and water bodies. In this chapter, to predict air quality in seven Indian smart cities, a data-driven artificial neural network (ANN) is merged with particle swarm optimization (PSO), a nature-inspired optimization technique. The PSO technique presented in this chapter functions as an ANN optimization technique. The data were obtained from the online web repository called Kaggle. The data-sets for 12 pollutants were selected for 7 Indian smart cities. The datasets were recorded during the last 5 years from 2016 to 2020 and were included for computation. Models can be used by air quality managers to forecast the effects of prospective new emissions. Policy makers often use models to predict ambient air pollution concentrations under various scenarios as a decision-making tool. Because of the robustness and effectiveness of AI models, the public will be able to get early warning for a high concentration level of pollutants in big cities, potentially lowering cardiovascular and respiratory mortality. The findings proved PSO's outstanding performance while evaluating high-dimensional data.
Elsevier, Advances in Computers, Volume 135, 2024, pp 129-154
