Evaluation of Data-Driven Methods for Hydrological Modeling: A Case Study of the Etobicoke Creek Watershed

In the past two decades, data-driven modeling has become a popular approach for different tasks. This paper presents an evaluation of performance five widely used approaches (i.e., generalized linear model, lasso regression, support vector machine, neural networks, and random forest) Etobicoke Creek watershed in Ontario, Canada. The models are built with eleven years meteorological hydrometric data from local stations, is examined by Nash-Sutcliffe efficiency coefficient, coefficient determination, mean absolute percentage error, root squared error. results show all able to generate acceptable predictions forest highest accuracy. study can provide selection hydrological future studies.