Analysis of various Machine Learning Regression Models for AQI Prediction

Because of the dramatic increase in particulate matter in the air, air quality prediction is a significant activity in many urban areas. The primary causes of air pollution are the combustion of fossil fuels, automobile and industrial emissions, and wildfires. The Air Quality Index (AQI) is a tool for determining the present level of air quality. Initially, the data set is obtained from Kaggle and pre-processed before training the model. Lasso Regression (Lasso R), Logistic Regression (LR), Decision Tree Regressor (DTR), Naive Bayes Regressor (NBR), K-Nearest Neighbor (KNN), and Random Forest Regressor (RFR) models are trained and tested using performance metrics like as accuracy, RMSE, MAE, MSE, and R2 score. According to the experimental data, RFR has a better accuracy (99%) and an excellent R2 score (0.998), as well as a low MAE (0.001), MSE (0.004), and RMSE (0.031) compared with other regression models