Development of Machine Learning Flood Model Using Artificial Neural Network (ANN) at Var River

Data-driven flow forecasting models, such as Artificial Neural Networks (ANNs), are increasingly used for operational flood warning systems. In this research, we systematically evaluate different machine learning techniques (random forest and decision tree) compare them with classical methods of the NAM rainfall run-off model Vésubie River, Nice, France. The modeled network is trained tested using discharge, precipitation, temperature, evapotranspiration data about four years (2011–2014). A comparative investigation executed to assess performance by Root Mean Squared Error (RMSE), Absolute (MAE), a correlation coefficient (R). According result, Feed Forward Network (FFNN) (a type ANN) models less efficient than models. precision parameters ANN 0.58 0.76 validation dataset. all tree which performed best had 0.99. prediction good compared training, opposite in model. can be improved fitting more input variables training dataset long period.