Groundwater Quality Assessment Based on Support Vector Machine 1

Abstract: Water quality assessment is a multivariate nonlinear system. Based on statistical learning theory, support vector machine (SVM) can transform the learning process into a convex quadratic planning problem to get a global optimization by using the rule of minimum structure risk, which is appropriate to solve small-sample, nonlinear classification and regression issues. Applying SVM in water quality assessment, the multiple-factor water quality assessment model based on SVM is established. According to groundwater quality assessment standard, water quality is divided into five grades. Eight assessment factors are selected to randomly generate sample set. All test samples are classified correctly after training the model. The model is applied for the assessment of karst groundwater sample at the Niangziguan fountain region of Haihe River basin to obtain the grade of water quality assessment. The result shows that such a method solves the complex nonlinear relationship between assessment factor and water quality grade. It offers high prediction accuracy and is a reasonable and feasible assessment method.