利用加拿大不列颠哥伦比亚省西南部QUEST南流沉积物地球化学数据的先进数据分析和机器学习进行矿产资源预测

IF 1 4区 地球科学 Q3 GEOCHEMISTRY & GEOPHYSICS Geochemistry-Exploration Environment Analysis Pub Date : 2020-10-09 DOI:10.1144/geochem2020-054
E. Grunsky, D. Arne
{"title":"利用加拿大不列颠哥伦比亚省西南部QUEST南流沉积物地球化学数据的先进数据分析和机器学习进行矿产资源预测","authors":"E. Grunsky, D. Arne","doi":"10.1144/geochem2020-054","DOIUrl":null,"url":null,"abstract":"In this study we apply multivariate statistical and predictive classification methods to interpret geochemical data from 8545 stream-sediment samples collected in southern British Columbia, Canada. Data for 35 elements were corrected for laboratory bias and adjusted for values reported below the lower limit of detection. Each sample site was attributed with the closest British Columbia MINFILE occurrence within 2.5 km. MINFILE occurrences were grouped into ‘GroupModels’ based on similarities between the British Columbia Geological Survey mineral deposit models and geochemical signatures. These data were used to create a training dataset of 474 observations, including 100 samples not attributed with a MINFILE occurrence. The training set was used to generate predictions for the mineral deposit models from which posterior probabilities were estimated for the remaining 8071 samples. The data underwent a centred log-ratio transformation and then characterization using either principal component analysis (PCA) or t-distributed stochastic neighbour embedding using 9 dimensions (t-SNE) prior to classification by random forests. The posterior probabilities generated from the t-SNE metric provide a slightly higher level of prediction accuracy compared to the posterior probabilities obtained using the PCA metric. The results are comparable to those obtained using a conventional catchment analysis approach and expert-driven model. The approach presented here provides a repeatable, consistent and defensible methodology for the identification of prospective mineralized terrains and mineral systems.","PeriodicalId":55114,"journal":{"name":"Geochemistry-Exploration Environment Analysis","volume":" ","pages":""},"PeriodicalIF":1.0000,"publicationDate":"2020-10-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://sci-hub-pdf.com/10.1144/geochem2020-054","citationCount":"15","resultStr":"{\"title\":\"Mineral-resource prediction using advanced data analytics and machine learning of the QUEST-South stream-sediment geochemical data, southwestern British Columbia, Canada\",\"authors\":\"E. Grunsky, D. Arne\",\"doi\":\"10.1144/geochem2020-054\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"In this study we apply multivariate statistical and predictive classification methods to interpret geochemical data from 8545 stream-sediment samples collected in southern British Columbia, Canada. Data for 35 elements were corrected for laboratory bias and adjusted for values reported below the lower limit of detection. Each sample site was attributed with the closest British Columbia MINFILE occurrence within 2.5 km. MINFILE occurrences were grouped into ‘GroupModels’ based on similarities between the British Columbia Geological Survey mineral deposit models and geochemical signatures. These data were used to create a training dataset of 474 observations, including 100 samples not attributed with a MINFILE occurrence. The training set was used to generate predictions for the mineral deposit models from which posterior probabilities were estimated for the remaining 8071 samples. The data underwent a centred log-ratio transformation and then characterization using either principal component analysis (PCA) or t-distributed stochastic neighbour embedding using 9 dimensions (t-SNE) prior to classification by random forests. The posterior probabilities generated from the t-SNE metric provide a slightly higher level of prediction accuracy compared to the posterior probabilities obtained using the PCA metric. The results are comparable to those obtained using a conventional catchment analysis approach and expert-driven model. The approach presented here provides a repeatable, consistent and defensible methodology for the identification of prospective mineralized terrains and mineral systems.\",\"PeriodicalId\":55114,\"journal\":{\"name\":\"Geochemistry-Exploration Environment Analysis\",\"volume\":\" \",\"pages\":\"\"},\"PeriodicalIF\":1.0000,\"publicationDate\":\"2020-10-09\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"https://sci-hub-pdf.com/10.1144/geochem2020-054\",\"citationCount\":\"15\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Geochemistry-Exploration Environment Analysis\",\"FirstCategoryId\":\"89\",\"ListUrlMain\":\"https://doi.org/10.1144/geochem2020-054\",\"RegionNum\":4,\"RegionCategory\":\"地球科学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q3\",\"JCRName\":\"GEOCHEMISTRY & GEOPHYSICS\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Geochemistry-Exploration Environment Analysis","FirstCategoryId":"89","ListUrlMain":"https://doi.org/10.1144/geochem2020-054","RegionNum":4,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"GEOCHEMISTRY & GEOPHYSICS","Score":null,"Total":0}
引用次数: 15

摘要

在这项研究中,我们应用多元统计和预测分类方法来解释加拿大不列颠哥伦比亚省南部8545个河流沉积物样本的地球化学数据。对35种元素的数据进行了实验室偏差校正,并对低于检测下限的报告值进行了调整。每个样本点的不列颠哥伦比亚省MINFILE发生率在2.5以内 根据不列颠哥伦比亚省地质调查局矿床模型和地球化学特征之间的相似性,将MINFILE矿点分为“GroupModels”。这些数据用于创建474个观测值的训练数据集,其中包括100个未归因于MINFILE事件的样本。训练集用于生成矿床模型的预测,从中估计剩余8071个样本的后验概率。数据经过中心对数比变换,然后在通过随机森林进行分类之前,使用主成分分析(PCA)或使用9维的t分布随机邻居嵌入(t-SNE)进行表征。与使用PCA度量获得的后验概率相比,从t-SNE度量生成的后验几率提供了略高水平的预测精度。结果与使用传统集水区分析方法和专家驱动模型获得的结果相当。本文提出的方法为识别潜在矿化地形和矿物系统提供了一种可重复、一致和可辩护的方法。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
查看原文
分享 分享
微信好友 朋友圈 QQ好友 复制链接
本刊更多论文
Mineral-resource prediction using advanced data analytics and machine learning of the QUEST-South stream-sediment geochemical data, southwestern British Columbia, Canada
In this study we apply multivariate statistical and predictive classification methods to interpret geochemical data from 8545 stream-sediment samples collected in southern British Columbia, Canada. Data for 35 elements were corrected for laboratory bias and adjusted for values reported below the lower limit of detection. Each sample site was attributed with the closest British Columbia MINFILE occurrence within 2.5 km. MINFILE occurrences were grouped into ‘GroupModels’ based on similarities between the British Columbia Geological Survey mineral deposit models and geochemical signatures. These data were used to create a training dataset of 474 observations, including 100 samples not attributed with a MINFILE occurrence. The training set was used to generate predictions for the mineral deposit models from which posterior probabilities were estimated for the remaining 8071 samples. The data underwent a centred log-ratio transformation and then characterization using either principal component analysis (PCA) or t-distributed stochastic neighbour embedding using 9 dimensions (t-SNE) prior to classification by random forests. The posterior probabilities generated from the t-SNE metric provide a slightly higher level of prediction accuracy compared to the posterior probabilities obtained using the PCA metric. The results are comparable to those obtained using a conventional catchment analysis approach and expert-driven model. The approach presented here provides a repeatable, consistent and defensible methodology for the identification of prospective mineralized terrains and mineral systems.
求助全文
通过发布文献求助,成功后即可免费获取论文全文。 去求助
来源期刊
Geochemistry-Exploration Environment Analysis
Geochemistry-Exploration Environment Analysis 地学-地球化学与地球物理
CiteScore
3.60
自引率
16.70%
发文量
30
审稿时长
1 months
期刊介绍: Geochemistry: Exploration, Environment, Analysis (GEEA) is a co-owned journal of the Geological Society of London and the Association of Applied Geochemists (AAG). GEEA focuses on mineral exploration using geochemistry; related fields also covered include geoanalysis, the development of methods and techniques used to analyse geochemical materials such as rocks, soils, sediments, waters and vegetation, and environmental issues associated with mining and source apportionment. GEEA is well-known for its thematic sets on hot topics and regularly publishes papers from the biennial International Applied Geochemistry Symposium (IAGS). Papers that seek to integrate geological, geochemical and geophysical methods of exploration are particularly welcome, as are those that concern geochemical mapping and those that comprise case histories. Given the many links between exploration and environmental geochemistry, the journal encourages the exchange of concepts and data; in particular, to differentiate various sources of elements. GEEA publishes research articles; discussion papers; book reviews; editorial content and thematic sets.
期刊最新文献
Multi-element geochemical analyses on ultrafine soils in Western Australia - Towards establishing abundance ranges in mineral exploration settings Alteration assemblage characterization using machine learning applied to high resolution drill-core images, hyperspectral data, and geochemistry Silver, cobalt and nickel mineralogy and geochemistry of shale ore in the sediment-hosted stratiform Nowa Sól Cu-Ag deposit, SW Poland Estimating the silica content and loss-on-ignition in the North American Soil Geochemical Landscapes datasets: a recursive inversion approach Spatial distribution, ecological risk and origin of soil heavy metals in Laoguanhe watershed of the Middle Route of China's South-to-North Water Diversion Project
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
现在去查看 取消
×
提示
确定
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
已复制链接
已复制链接
快去分享给好友吧!
我知道了
×
扫码分享
扫码分享
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术
文献互助 智能选刊 最新文献 互助须知 联系我们:info@booksci.cn
Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。
Copyright © 2023 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1