The purpose of this research is to assess the efficacy of five distinct artificial intelligence model techniques: AdaBoost, Gradient Boosting, Tree, Random Forest, and KNN, to estimate the water quality parameters of dissolved oxygen (DO) and biochemical oxygen demand (BOD). The performance of each model was assessed using two datasets: Al-Muthanna Bridge and Al-Aammah Bridge on the Tigris River in Baghdad City. The data was randomly divided into two categories: 70% for training and 30% for testing. Principal component analysis (PCA) was used to identify the most effective input parameters for modeling DO and BOD. The four performance criteria—coefficient of determination (R2), root mean square error (RMSE), mean absolute error (MAE), and mean square error (MSE)— were applied in order to evaluate the models' effectiveness. It was demonstrated that the AdaBoost and Gradient Boosting models were superior for predicting DO and BOD. For DO prediction, the coefficient of determination R2 of Gradient Boosting (AdaBoost) at Al-Muthanna Bridge and Al-Aammah Bridge were 0.994 (0.992) and 0.994 (0.991), respectively. For BOD prediction, the correlation coefficients R2 of Gradient Boosting (AdaBoost) were 0.992 (0.982) and 0.989 (0.990), respectively. This study has shown that sophisticated machine learning techniques, such as gradient boosting and AdaBoost, are suitable for predicting water quality indices. They could also be helpful for predicting and managing the water quality parameters of different water supply systems in the future in water-related communities where artificial intelligence technology is still being thoroughly investigated.