机器学习在岩矿地球化学研究中的应用——综述与思考  

A review on the machine learning approach to rockand mineral geochemistry research

作　　者：谢玉芝 汪洋[1] XIE Yuzhi;WANG Yang(School of Earth Science and Resources,China University of Geosciences(Beijing),Beijing,100083)

机构地区：[1]中国地质大学(北京)地球科学与资源学院,北京100083

出　　处：《地质论评》2023年第4期1465-1474,共10页Geological Review

基　　金：中国地质调查局地质调查项目(编号:D1912)的成果~~。

摘　　要：岩石与矿物的地球化学成分数据具有高维度特征。传统的岩矿地球化学成分研究主要采用二元/三元图解判别法,准确率不高,在数理统计方法上有欠缺。机器学习方法非常适用于对大样本高维度的岩矿成分数据进行数理统计处理。笔者等在介绍机器学习常见算法基本原理的基础上,总结近5年来国内外学者将机器学习方法应用于岩石矿物成分数据研究的实例,包括:①根据矿物成分溯源其母岩(源岩)、判别矿床类型,②新生代火山岩溯源,③判别变质岩原岩,④依据岩浆岩成分判别大地构造环境等。已有的研究实例显示,机器学习方法的准确度明显优于传统的低维度判别法。机器学习本质是分析大样本数据的高维度变量之间的相关、归类等多元统计问题。推广机器学习的应用需要建设开放获取(Open Access)的矿物、岩石成分数据库,同时全面实施开放研究(Open Research)的发表策略.The geochemical composition data of rocks and minerals have high dimensional characteristics.The conventional study on the geochemical composition of rocks and minerals mainly adopts the binary/ternary graphical discrimination method,which has low accuracy and lacks solid basis of mathematical statistics.The machine learning method is very suitable for the statistical processing of large scale high dimensional data as the rock and mineral compositions.On the basis of introducing the basic principles of common machine learning algorithms,this paper summarizes the case studies of the machine learning approach to the rock and mineral geochemistry in the past five years,including:①discrimination of the source rock of the minerals from their compositions,②distinguishing the type of deposit from the mineral compositions,③identifying the provenance of Cenozoic volcanic rocks,④distinguishing the proto-lithology discrimination for metamorphic rocks,and⑤tectonic discrimination of magmatic rocks,etc.Compared with conventional low-dimensional discrimination method,the machine learning approach provides higher accuracy and the ability to process the high-dimensional data.The nature of machine learning approach is to perform the multivariate statistical analysis,such as the correlation and classification among the high-dimensional variables of large sample data.For popularizing the machine-learning approach in petrological community,more open accessed databases of mineral and rock compositions are needed,and the Open Research policy should be fully implemented in academic publications.

关 键 词：机器学习 岩矿成分 判别图解 大数据 

分 类 号：P632[天文地球—地质矿产勘探]