Improving the Compound Channel Discharge Prediction model byJB's method.

In hydrological engineering, discharge prediction in compound channels is important forflood control and river structure. This prediction is investigated in the classical method byML models.However, more suitable methods were presented later to improve this prediction, which isknown as machine learning methods. Among these methods, we can mention the AdaptiveNeural Fuzzy Inference System (ANFIS) and Support Vector Machine (SVM). Bootstrap as aresampling technique on the above methods improved the presented models in prediction.In this paper, the Jackknife-After-Bootstrap (JB) method is proposed as a resampling method,which is effective in the predictive power of the model and its performance

کلیدواژه ها:

ANFIS ، Bootstrap ، JB ، RMSE ، 𝑅۲ ، SVM.