Prediction-based approaches are valuable in assessing dam safeties, as they allow comparing the actual measurements with the projected values to detect anomalies early. For two decades, machine learning (ML) algorithms have been developed and improved to help in accurately predicting the dam behaviors. However, the generalization ability (GA) of these models is not analyzed enough in dam engineering. In this study, the Multiple Linear Regression (MLR), Artificial Neural Network (ANN), Support Vector Regression (SVR), and Adaptive Boosting (AdaBoost) models with nonlinear autoregressive exogenous inputs (NARX) are evaluated and compared with the conventional Hydrostatic Seasonal Time (HST) model for predicting the daily pore water pressure in an embankment Dam. Moreover, we proposed a classification method of the model into four categories ‘’Perfect’’, ‘’Excellent’’, ‘’Good’’, and ‘’Poor’’ according to the GA. Results showed that, except for the AdaBoost, the other ML models outperformed the traditional statistical approach (HST) in terms of prediction accuracy as well as the GA. Overall; the study results provide new insights in enhancing the monitoring processes and dam safeties by detecting the anomalies early through the measurements and the selection of the best fitted-models.