| 王乔.随机森林在水系沉积物微量元素含量预测中的应用——以山西省关帝山一带为例[J].矿产勘查,2025,16(S1):197-200 |
| 随机森林在水系沉积物微量元素含量预测中的应用——以山西省关帝山一带为例 |
| Application of random forest to the prediction of trace element content in stream sediment:Taking the Guandishan Area of Shanxi Province as an example |
| |
| DOI：10.20008/j.kckc.2025S1023 |
| 中文关键词: 机器学习 水系沉积物 微量元素含量预测 关帝山 |
| 英文关键词: machine learning stream sediment prediction of trace element content Guandishan Area |
| 基金项目: |
|
| 摘要点击次数: 480 |
| 全文下载次数: 156 |
| 中文摘要: |
| 本文利用机器学习方法随机森林，依托关帝山一带 1∶20万化探数据，选取主量元素 SiO2、Al2O3、CaO、 Fe2O3、K2O、MgO、Na2O、Mn、P、Ti作为特征元素，预测关帝山一带水系沉积物微量元素的含量。发现关帝山一带水系沉积物主量元素与微量元素含量之间具有高维的定量关系，本方法对地质系统数据收集和预测补全具有广阔的应用推广前景。 |
| 英文摘要: |
| In this study，we employed the random forest machine learning algorithm to predict trace elementconcentrations in stream sediments within the Guandishan area. Utilizing 1∶200，000 scale geochemical exploration data，we selected SiO2，Al2O3，CaO，Fe2O3，K2O，MgO，Na2O，Mn，P，and Ti as feature elements for the model. Our analysis revealed a significant high-dimensional quantitative relationship between the concentrations of majorelements and trace elements in the stream sediments of the Guandishan region. This methodology demonstratesconsiderable potential for application in geological data collection and predictive modeling，offering a promising approach for enhancing the completeness and accuracy of geochemical datasets. |
|
查看全文
查看/发表评论 下载PDF阅读器 |
| 关闭 |