Air Quality Prediction (Meteorology)

Explored pollutant–weather correlations using heatmaps and distributions; performed preprocessing using imputation, scaling, and one-hot encoding. Trained Linear Regression, SVR, Random Forest, Gradient Boosting, and LightGBM using GridSearchCV and cross-validation. Evaluated using R²/MAE/MSE/RMSE and checked over/underfitting using train vs. validation errors comparison; visualized results in clean tables/plots.