Advancing Air Quality Index (AQI) Forecasting: Traditional, Data-Driven, and Hybrid Techniques

The Air Quality Index (AQI) is a precise instrument for assessing the intensity of air pollution, which is a major adverse weather condition for the environment and human health these days. Public health involvement and the implementation of warning systems are crucial for accurate AQI forecasts. This work provides an overview of the various algorithms used for AQI forecasting, from traditional methods to the latest Machine Learning (ML) Methods, and hybrid models. This analysis focuses on the strengths and limitations of different approaches in detail including Auto-Regressive Integrated Moving Average (ARIMA), Support Vector Machine (SVM), Random Forest (RF), Extreme Learning Machines (ELM), and hybrid models like Genetic Algorithms -Extreme Learning Machines (GA-ELM) and Long Short-Term Memory-Convolutional Neural Network (LSTM-CNN). This work also explained optimizing techniques like GA and Particle Swarm Optimization (PSO). Finally concluded with an explanation of future directions in AQI forecasting.