论文摘要
The random forest algorithm is a nonlinear model that can map complex functional relations. It has been widely applied in many fields. This paper studies the application of random forest algorithm in geochemical anomaly identification. The random forest algorithm can be used to fit the complex relationship among various metallogenic elements and provide efficient decision-making tools for mineral prediction. This paper takes the geochemical data of the akille mining area in east WuZhuQi,the Inner Mongolia Autonomous Region,as the training sample and the test sample. The random forest algorithm is used to deal with the samples. The test shows that the convergence rate of the random forest method is faster and no over fitting phenomenon is produced,and the geochemical exploration can be made on the premise of superposition of the ore point map. The data are classified by anomaly and background,and the accuracy of the classification is up to 89%. This method also obtains the prediction model of geochemical elements combination in this area, which is practical and improves the precision of mineral prediction.
论文目录
文章来源
类型: 国际会议
作者: CHEN Zhen,CHEN Jianping,AN Zhihong
来源: 第九届世界华人地质科学研讨会 2019-06-01
年度: 2019
分类: 基础科学,工程科技Ⅰ辑
专业: 地质学,矿业工程
单位: School of Earth Sciences & Resources, China University of GeosciencesChina Aero Geophysical Survey and Remote Sensing Center for Land and Resources
分类号: P632
DOI: 10.26914/c.cnkihy.2019.028501
页码: 399
总页数: 1
文件大小: 94k
下载量: 2