River discharge prediction using Acoustic Doppler Current Profiler (ADCP) data entails utilizing data obtained from ADCP sensors for predicting the total volume of water flowing through a water body at an identifiable location at a given time. ADCP is a hydroacoustic device that utilizes sound waves to calculate the individual velocities of water particles at varying depths to explain the complexities of river dynamics. In this study, ML models including Logistic Regression (LR), Random Forest (RF), Support Vector Classification (SVC), eXtreme Gradient Boosting (XGB), Light Gradient Boosting Machine (LGBM), K-Nearest Neighbors (KNN), Decision Tree Classifier (DTC), and Multi-Layer Perceptron (MLP), were employed and evaluated according to key performance indicators. The ML models were evaluated based on these metrics: Precision, Recall, F-1 score, and Area Under the Receiver Operating Characteristic Curve (AUC ROC). Overall, the results show that the RF model performed excellently with a 0.977 accuracy, 0.977 precision, 0.995 AUC ROC, 0.977 F-1 score, and 0.977 recall. The LGBM model next showed comparable results in precision and accuracy but with a slight increase in F-1 score and AUC ROC metrics. In contrast, in the performance metrics of LR, SVC, XGB, and MLP the accuracies range from 0.943 to 0.977. The DTC had a strong F-1 score of 0.967 demonstrating its ability to capture both positive and negative examples accurately. The success of ML in estimating river discharge from ADCP data demonstrates the importance of sophisticated algorithms in hydrological modeling.