Evaluation of the success of the hybrid wavelet-based ANFIS approach in the estimation of monthly stream flows of the Bitlis River, Turkey

Abstract This study aimed to predict monthly flows using an adaptive neuro-fuzzy inference system (ANFIS) and wavelet-ANFIS (W-ANFIS) determine the effect of wavelet transformation on success machine learning model. For this purpose, model inputs are divided into three subcomponents with Daubechies 10 mother wavelets. Subcomponents highest correlation were chosen as inputs. The most suitable models selected by dividing 3–7 sub-sets, 11 different lagged input combinations, testing various membership functions. In establishing ANFIS model, 75% data used for training 25% testing. performance was evaluated root mean square error, Pearson coefficient, determination coefficients, Taylor diagram. A two a hybrid algorithm, Gbellmf function, 400 iterations suitable. It concluded that W-ANFIS transform method increased established Moreover, it suggested in can be effectively similar climatic regions fed snowmelt dominated semi-arid climate.