Kiyoumars Roushangar - Associate Professor, Department of Civil Engineering, University of Tabriz, Tabriz, Iran,
Roghayeh Ghasempour - PH.D student, Department of Civil Engineering, University of Tabriz, Tabriz, Iran,
Hassan Sani - M.Sc, Department of Civil Engineering, University of Tabriz, Tabriz, Iran,,Member Yong Researchers and Elit Club
Farhad Alizade Afshar - M.Sc ,Civil Engineering

An accurate estimation of flow discharge is of great importance due to its significant effects on reducing the damaging impacts of floods. Numerous of studies have been done about flood prediction process. In this study, the performance of the Support Vector Machine (SVM) and Adapted Neural Fuzzy Inference System (ANFIS) as Meta model approaches were assessed in river discharge prediction. Statistical error criteria were used for evaluating the accuracy of the models. According to the obtained results, the SVM and ANFIS models were found to be reliable in flood prediction. The results indicated that the impact of Q (river discharge) is more than H (river flow level) in flood discharge prediction process.

ANFIS, Discharge, River, SVM