SVM based ore grade valuation model construction

Considering the excellent convex optimization property and the good ability of generalization of support vector machines (SVM), this paper gives out an grade valuation model based on it. Through the processing of data normalization and the optimization analysis of some factor, the SVM based ore grade valuation model is created. The prediction results compare with the Thiessen polygons method, the Distance power inverse ratio method and the Kriging interpolation method, which verifies the feasibility and validity of the SVM based ore grade valuation model.