Modeling and Application of Ore Grade Interpolation Based on SVM
Support Vector Machine (SVM) has become an effective machine learning method characterized by solving learning problems of small samples, nonlinearity and high-dimensional pattern recognition. Based on Support Vector Machine Regression (SVR), the paper presents an ore grade interpolation model by using the cross-validation contrast to select the kernel function and the model parameters including penalty parameter C, the insensitive coefficient e and the kernel function parameter 0. Then the model is applied in a typical domestic underground mine and the interpolation result shows the model is feasible and more efficient in contrast with the production data and the results of traditional interpolation methods, such as the Thiessen polygon method, the distance power inverse ratio method and the Kriging interpolation method.
Support Vector Machine ore grade interpolation mine kernalfunction model
Cuiping Li Yaoxia Zheng Zhongxue Li Yiqing Zhao
State Key Laboratory of High-efficient Mining and Safety of Metal Mines, Ministry of Education University of Science and Technology Beijing Beijing 100083, China
国际会议
上海
英文
1570-1573
2011-07-26（万方平台首次上网日期，不代表论文的发表时间）