Prediction of channel sinuosity in perennial rivers using Bayesian Mutual Information theory and support vector regression coupled with meta-heuristic algorithms

Support Vector Regression (SVR) combined with Invasive Weeds Optimization (IWO), standalone SVR, and Radial Basis Function Neural Networks are applied to estimate channel sinuosity in perennial rivers. With this aim, a dataset 132 data related geomorphologic data, corresponding 119 streams, is considered. Bayesian Mutual Information theory used determine the parameters affecting reveal that bankfull depth affects most. Seven input parameter combinations for prediction considered, both training testing stages, SVR-IWO model $$\left( {R_{Train} = 0.959,RMSE_{Train} 0.072, MAE_{Train} 0.037, R_{test} 0.892, RMSE_{Test} 0.103, MAE_{Test} 0.065} \right)$$ R Train 0.959 , M S E 0.072 A 0.037 test 0.892 Test 0.103 0.065 shows best performance while SVR generated results performances of 0.792,RMSE_{Train} 0.158, 0.141, 0.704, 0.163, 0.151} 0.792 0.158 0.141 0.704 0.163 0.151 . Model uncertainty quantified terms entropy three models further confirming set predicted by closest observed set.