Sihag, Parveen
ESTIMATION OF CUMULATIVE INFILTRATION OF SOIL USING FIELD - Vol.39(1), Jan - Roorkee Indian Water Resources Society 2019 - 14-22p.
.cumulative infiltration, Support vector regression, Gaussian process regression, Adaptive neuro-fuzzy inference system.13 system have been used as dominant tools in solving water Parsaie and Haghiabi, 2014; Parsaie and Haghiabi, 2015; Azamathulla et al., 2016; Tiwari et al., ag et al., 2018b; Tiwari et al., 2018). The advantages of using SVM, GP and ANFIS are that these defined parameters. Keeping in view of the improved performance by SVM, GP and ANFIS approaches in water engineering problems; this study compares its performances with empirical models (Kostiakov model and SCS model) of cumulative Support vector Machines This method was introduced by Vapnik (1995) and derived from statistical learning theory. Main principle of SVM is optimal separation of classes, from the separable classes SVM selects the one which have least generalisation error from infinite number of linear classifier or set upper limit to error which is obtained from structural risk minimisation. Thus maximum margin between two classes could be obtained from the selected hyper plane and sum of distances of the hyper plane from the closest point of two classes will een two classes. For further more details study of SVM readers are referred to (Vapnik (1995) An Overview of Gaussian Process Regression (GP) (2006) assumption on which GP regression model works are that adjoining observation should express information about each other, it is a method of specifying a prior directly over function space. Mean and covariance of Gaussian distribution is vector and matrix J. Indian Water Resour.
Civil Engineering
ESTIMATION OF CUMULATIVE INFILTRATION OF SOIL USING FIELD - Vol.39(1), Jan - Roorkee Indian Water Resources Society 2019 - 14-22p.
.cumulative infiltration, Support vector regression, Gaussian process regression, Adaptive neuro-fuzzy inference system.13 system have been used as dominant tools in solving water Parsaie and Haghiabi, 2014; Parsaie and Haghiabi, 2015; Azamathulla et al., 2016; Tiwari et al., ag et al., 2018b; Tiwari et al., 2018). The advantages of using SVM, GP and ANFIS are that these defined parameters. Keeping in view of the improved performance by SVM, GP and ANFIS approaches in water engineering problems; this study compares its performances with empirical models (Kostiakov model and SCS model) of cumulative Support vector Machines This method was introduced by Vapnik (1995) and derived from statistical learning theory. Main principle of SVM is optimal separation of classes, from the separable classes SVM selects the one which have least generalisation error from infinite number of linear classifier or set upper limit to error which is obtained from structural risk minimisation. Thus maximum margin between two classes could be obtained from the selected hyper plane and sum of distances of the hyper plane from the closest point of two classes will een two classes. For further more details study of SVM readers are referred to (Vapnik (1995) An Overview of Gaussian Process Regression (GP) (2006) assumption on which GP regression model works are that adjoining observation should express information about each other, it is a method of specifying a prior directly over function space. Mean and covariance of Gaussian distribution is vector and matrix J. Indian Water Resour.
Civil Engineering