基于机器学习的锆石成因分类研究

doi:10.13745/j.esf.sf.2022.2.75

地学前缘 ›› 2022, Vol. 29 ›› Issue (5): 464-475.DOI: 10.13745/j.esf.sf.2022.2.75

基于机器学习的锆石成因分类研究

朱紫怡¹(), 周飞¹, 王瑀¹, 周统¹, 侯照亮², 邱昆峰¹^,³^,^*()

1.中国地质大学(北京) 地球科学与资源学院, 北京 100083
2.奥地利维也纳大学地质系, 维也纳 1090
3.中国地质大学地质过程与矿产资源国家重点实验室, 北京 100083

收稿日期:2021-12-16 修回日期:2022-03-18 出版日期:2022-09-25 发布日期:2022-08-24
通信作者: 邱昆峰
作者简介:朱紫怡(2000—),女,本科生,地质学专业。E-mail: ziyizhuuu@qq.com
基金资助:
国家自然科学基金项目(91962106);国家自然科学基金项目(42111530124);国家自然科学基金项目(42072087);国家重点研发计划项目(2019YFA0708603);中国高校产学研创新基金资助课题(2021ALA01006);111计划2.0(BP0719021);中国地质大学(北京)创新创业训练计划项目(S202111415004)

Machine learning-based approach for zircon classification and genesis determination

ZHU Ziyi¹(), ZHOU Fei¹, WANG Yu¹, ZHOU Tong¹, HOU Zhaoliang², QIU Kunfeng¹^,³^,^*()

1. School of Earth Sciences and Resources, China University of Geosciences (Beijing), Beijing 100083, China
2. Department of Geology, University of Vienna, Vienna 1090, Austria
3. State Key Laboratory of Geological Process and Mineral Resources, China University of Geosciences (Beijing), Beijing 100083, China

Received:2021-12-16 Revised:2022-03-18 Online:2022-09-25 Published:2022-08-24
Contact: QIU Kunfeng

摘要/Abstract

摘要：

锆石是在自然界中多种温压条件下能够稳定保存,并记录原岩年龄信息的副矿物。锆石微量元素能完整记录地质演化过程信息。通过微量元素分析锆石成因的研究已久,通常利用Th-U图解和La_N-(Sm/La)_N图解等二元图解对锆石进行分类研究。然而,随着锆石研究的深入,以及二元图解无法呈现数据高维度信息的局限性,传统图解已经不能满足对锆石类型进行准确判别,且对已知类型的锆石出现判定偏差。因此,本文将地质大数据与机器学习相结合,训练出高维度锆石成因分类器。文中收集了3 498条不同成因类型的锆石微量元素数据,并通过测试和运用随机森林、支持向量机、人工神经网络和k近邻等4种机器学习算法,最终得出准确率为86.8%的线性支持向量机锆石成因分类器,用于锆石类型的判定与预测。这项工作为锆石分类研究提供了更高维度的判别手段,极大提高了微量元素分析成因结果的精度。将锆石微量元素数据与机器学习方法相结合,是大数据分析与机器学习技术在地球化学研究中的积极探索。

关键词: 锆石, 微量元素, 成因, 大数据分析, 机器学习

Abstract:

Zircon, a stable paragenetic mineral in various geological environments, has been recognized as a great tool in the chronological study of primary rocks. Trace elements in zircons may reveal geological evolutionary processes, and have long been used in zircon classification and zircon formation studies by binary diagram method, such as Th-U and La_N-(Sm/La)_N diagrams. However, with the massive increase of zircon research, the traditional binary diagrams are no longer adequate for a precise determination of zircon types because binary plots cannot display higher dimensional information and therefore can lead to erroneous interpretation of zircon data. To address this issue, we take a machine learning approach to analyzing 3 498 zircon trace element data for different zircon genetic types to obtain high-dimensional zircon classification diagrams. We tested four machine learning algorithms (random forest, support vector machine, artificial neural network, and k-nearest neighbor) and consider support vector machine, with an 86.8% accuracy in predicting zircon type and origin, can best contribute to zircon classification. In addition to the development of a high-dimensional zircon classification diagram, this work also greatly improves the accuracy of zircon genesis determination using trace elements, and demonstrates the applicability of modern data science technique in geochemical research.

Key words: zircon, trace elements, formation, big data analysis, machine learning

中图分类号:

P581

朱紫怡, 周飞, 王瑀, 周统, 侯照亮, 邱昆峰. 基于机器学习的锆石成因分类研究[J]. 地学前缘, 2022, 29(5): 464-475.

ZHU Ziyi, ZHOU Fei, WANG Yu, ZHOU Tong, HOU Zhaoliang, QIU Kunfeng. Machine learning-based approach for zircon classification and genesis determination[J]. Earth Science Frontiers, 2022, 29(5): 464-475.

图/表 12

图1 不同类型的锆石微量元素散点图(底图区域据文献[9])

Fig.1 Scatter plot of the published zircon-associated trace elements from various rock types. Background region based on [9].

表1 已发表的锆石成因类型判别图解准确率

Table 1 The accuracy of the published diagram to classify different zircon provenance

判别图解	准确率/%
判别图解	岩浆锆石	热液锆石	变质锆石	平均
Th/U	81.65	-	23.5	73.1
La_N-(Sm/La)_N	42.4	20.9	-	38.5

表2 不同母岩类型锆石微量元素数据量

Table 2 Published data of zircon trace elements from different provenance

序号	母岩类型	数据量	参考文献
1	岩浆锆石	2466	[9,25⇓⇓⇓⇓⇓⇓⇓⇓⇓⇓⇓⇓⇓⇓⇓⇓⇓⇓⇓⇓-46]
2	热液锆石	508	[9,30,35,37-38,41-42,47⇓⇓⇓⇓⇓⇓⇓-55]
3	变质锆石	524	[33-34,49,56⇓⇓⇓⇓⇓-62]

图2 本文所收集数据在全球分布情况

Fig.2 Global distribution of the collected zircon data

图3 锆石微量元素数据含量箱形图

Fig.3 Trace element concentrations of the various zircons

表3 不同母岩类型锆石微量元素数据量缺失情况

Table 3 Missing data of zircon trace elements from different provenance

元素	元素数据缺失率			元素	元素数据缺失率
元素	热液锆石	岩浆锆石	变质锆石	元素	热液锆石	岩浆锆石	变质锆石
La	0%	7%	23%	Tm	4%	1%	17%
Ce	0%	0%	0%	Yb	0%	0%	0%
Pr	0%	1%	4%	Lu	0%	0%	12%
Nd	0%	0%	2%	Y	20%	7%	38%
Sm	0%	0%	1%	Hf	27%	8%	41%
Eu	1%	0%	1%	Nb	38%	23%	58%
Gd	0%	0%	0%	Ta	36%	67%	65%
Tb	5%	1%	20%	Ti	40%	23%	47%
Dy	0%	0%	0%	Pb	42%	79%	78%
Ho	5%	0%	14%	Th	13%	2%	15%
Er	0%	0%	2%	U	13%	2%	15%

图4 该数据集中锆石穷举投图排名前9的二维图解

Fig.4 Top 9 biplots from the dataset of zircon by using exhaustive method

表4 4种模型的最优超参数和在测试集的准确率

Table 4 Hyper-parameters of the 4 models and the accuracy on the test set

模型	准确率	最优超参数
SVM(linear)	0.868	C = 16
SVM(RBF)	0.802	C=16,gamma = 0.5
KNN	0.860	weights = distance, n neighbors = 1
ANN	0.899	alpha = 0.001, activation = tanh, hidden layer sizes = (20,1)
Random Forest	0.878	n estimators = 100, max depth = 30, max features = 16

图5 混淆矩阵基于线性内核支持向量机的分类器的验证结果

Fig.5 Confusion matrix showing results of linear SVM classifier

表5 测试集在支持向量机模型中的精确率、召回率、F1分数

Table 5 Accuracy, recall, and F1 score of the test set in the SVM

锆石类型	精确率	召回率	F1分数	数量
热液锆石	0.855	0.855	0.855	76
岩浆锆石	0.857	0.868	0.863	76
变质锆石	0.893	0.882	0.887	76
宏平均(macro)			0.868	228

图6 不同算法对于特征构建和过采样的反应

Fig.6 Responses of different algorithms to feature construction and oversampling

图7 混淆矩阵基于4种分类器的验证结果

Fig.7 Confusion matrix showing results of the four classifiers

参考文献 75

[1]	陈道公, 李彬贤, 夏群科, 等. 变质岩中锆石U-Pb计时问题评述: 兼论大别造山带锆石定年[J]. 岩石学报, 2001, 17 (1): 129-138.
[2]	李长民. 锆石成因矿物学与锆石微区定年综述[J]. 地质调查与研究, 2009, 32(3): 161-174.
[3]	VAVRA G, SCHMID R, GEBAUER D. Internal morphology, habit and U-Th-Pb microanalysis of amphibolite-to-granulite facies zircons: geochronology of the Ivrea Zone (Southern Alps)[J]. Contributions to Mineralogy and Petrology, 1999, 134(4): 380-404. DOI URL
[4]	VAVRA G, GEBAUER D, SCHMID R, et al. Multiple zircon growth and recrystallization during polyphase Late Carboniferous to Triassic metamorphism in granulites of the Ivrea Zone (Southern Alps): an ion microprobe (SHRIMP) study[J]. Contributions to Mineralogy and Petrology, 1996, 122(4): 337-358. DOI URL
[5]	MEZGER K, KROGSTAD E J. Interpretation of discordant U-Pb zircon ages: an evaluation[J]. Journal of Metamorphic Geology, 1997, 15(1): 127-140. DOI URL
[6]	吴元保, 郑永飞. 锆石成因矿物学研究及其对U-Pb年龄解释的制约[J]. 科学通报, 2004, 49(16): 1589-1604.
[7]	HACKER B R, RATSCHBACHER L, WEBB L, et al. U/Pb zircon ages constrain the architecture of the ultrahigh-pressure Qinling-Dabie Orogen, China[J]. Earth and Planetary Science Letters, 1998, 161(1/2/3/4): 215-230. DOI URL
[8]	HERMANN J, RUBATTO D, KORSAKOV A, et al. Multiple zircon growth during fast exhumation of diamondiferous, deeply subducted continental crust (Kokchetav Massif, Kazakhstan)[J]. Contributions to Mineralogy and Petrology, 2001, 141(1): 66-82. DOI URL
[9]	HOSKIN P W O. Trace-element composition of hydrothermal zircon and the alteration of Hadean zircon from the Jack Hills, Australia[J]. Geochimica et Cosmochimica Acta, 2005, 69(3): 637-648. DOI URL
[10]	雷玮琰, 施光海, 刘迎新. 不同成因锆石的微量元素特征研究进展[J]. 地学前缘, 2013, 20(4): 273-284.
[11]	周永章, 王俊, 左仁广, 等. 地质领域机器学习、深度学习及实现语言[J]. 岩石学报, 2018, 34(11): 3173-3178.
[12]	RECANATI A, GROZAVU N, BENNANI Y, et al. Apatite (U-Th-Sm)/He date dispersion: first insights from machine learning algorithms[J]. Earth and Planetary Science Letters, 2021, 554: 116655. DOI URL
[13]	WANG Y, QIU K F, MÜLLER A, et al. Machine learning prediction of quartz forming-environments[J]. Journal of Geophysical Research: Solid Earth, 2021, 126(8): 1-11.
[14]	黄发明, 潘李含, 姚池, 等. 基于半监督机器学习的滑坡易发性预测建模[J]. 浙江大学学报(工学版), 2021, 55(9): 1705-1713.
[15]	GUO P, YANG T, XU W L, et al. Machine learning reveals source compositions of intraplate basaltic rocks[J]. Geochemistry Geophysics Geosystems, 2021, 22(9): e2021GC009946.
[16]	ZHONG R C, DENG Y, LI W B, et al. Revealing the multi-stage ore-forming history of a mineral deposit using pyrite geochemistry and machine learning-based data interpretation[J]. Ore Geology Reviews, 2021, 133: 104079. DOI URL
[17]	周永章, 黎培兴, 王树功, 等. 矿床大数据及智能矿床模型研究背景与进展[J]. 矿物岩石地球化学通报, 2017, 36(2): 327-331, 344.
[18]	周志华. 机器学习[J]. 中国民商, 2016 (3): 93.
[19]	王瑀, 邱昆峰, 侯照亮, 等. 石英Ti/Ge-P: 基于机器学习的矿床类型判别新图解[J]. 岩石学报, 2022, 38(1): 281-290.
[20]	周统, 邱昆峰, 王瑀, 等. 磷灰石Eu/Y-Ce: 基于大数据的源区类型判别新图解[J]. 岩石学报, 2022, 38(1): 291-299.
[21]	周子扬. 机器学习与深度学习的发展及应用[J]. 电子世界, 2017 (23): 72-73.
[22]	SAMUEL A L. Some studies in machine learning using the game of checkers[J]. IBM Journal of Research and Development, 2000, 44(1/2): 206-226. DOI URL
[23]	耿厅, 周永章, 李兴远, 等. 锆石微量元素对成矿岩体的判别: 来自大数据思维的应用[J]. 地质通报, 2019, 38(12): 1992-1998.
[24]	周永章, 左仁广, 刘刚, 等. 数学地球科学跨越发展的十年: 大数据、人工智能算法正在改变地质学[J]. 矿物岩石地球化学通报, 2021, 40(3): 556-573, 777.
[25]	HARTMANN L A, BAGGIO S B, BRÜCKMANN M P, et al. U-Pb geochronology of Paranávolcanics combined with trace element geochemistry of the zircon crystals and zircon Hf isotope data[J]. Journal of South American Earth Sciences, 2019, 89: 219-226. DOI URL
[26]	GUO L, ZHANG H F, HARRIS N, et al. Detrital zircon U-Pb geochronology, trace-element and Hf isotope geochemistry of the metasedimentary rocks in the Eastern Himalayan syntaxis: tectonic and paleogeographic implications[J]. Gondwana Research, 2017, 41: 207-221 DOI URL
[27]	YAN L L, HE Z Y, BEIER C, et al. Zircon trace element constrains on the link between volcanism and plutonism in SE China[J]. Lithos, 2018, 320/321: 28-34. DOI URL
[28]	YAN L L, HE Z Y, KLEMD R, et al. Tracking crystal-melt segregation and magma recharge using zircon trace element data[J]. Chemical Geology, 2020, 542: 119596. DOI URL
[29]	MELO M G, LANA C, STEVENS G, et al. Assessing the isotopic evolution of S-type granites of the Carlos Chagas Batholith, SE Brazil: clues from U-Pb, Hf isotopes, Ti geothermometry and trace element composition of zircon[J]. Lithos, 2017, 284/285: 730-750. DOI URL
[30]	YUAN F, LIU J J, CARRANZA E J M, et al. Zircon trace element and isotopic (Sr, Nd, Hf, Pb) effects of assimilation-fractional crystallization of pegmatite magma: a case study of the Guangshigou biotite pegmatites from the North Qinling Orogen, central China[J]. Lithos, 2018, 302/303: 20-36. DOI URL
[31]	TURLIN F, VANDERHAEGHE O, GERVAIS F, et al. Petrogenesis of LREE-rich pegmatitic granite dykes in the central Grenville Province by partial melting of Paleoproterozoic-Archean metasedimentary rocks: evidence from zircon U-Pb-Hf-O isotope and trace element analyses[J]. Precambrian Research, 2019, 327: 327-360. DOI URL
[32]	WANG Q, ZHU D C, ZHAO Z D, et al. Magmatic zircons from I-, S- and A-type granitoids in Tibet: trace element characteristics and their application to detrital zircon provenance study[J]. Journal of Asian Earth Sciences, 2012, 53: 59-66. DOI URL
[33]	DONG X, ZHANG Z M, NIU Y L, et al. Reworked Precambrian metamorphic basement of the Lhasa terrane, southern Tibet: Zircon/titanite U-Pb geochronology, Hf isotope and geochemistry[J]. Precambrian Research, 2020, 336: 105496. DOI URL
[34]	ZHENG J P, GRIFFIN W L, O'REILLY S Y, et al. Neoarchean (2.7-2.8 Ga) accretion beneath the North China Craton: U-Pb age, trace elements and Hf isotopes of zircons in diamondiferous kimberlites[J]. Lithos, 2009, 112(3/4): 188-202. DOI URL
[35]	BORBA M L, TASSINARI C C G, MATOS F M V, et al. Tracking hydrothermal events using zircon REE geochemistry from the Carajás Mineral Province, Brazil[J]. Journal of Geochemical Exploration, 2021, 221: 106679. DOI URL
[36]	张小文, 向华, 钟增球, 等. 海南尖峰岭岩体热液锆石U-Pb定年及微量元素研究: 对热液作用及抱伦金矿成矿时代的限定[J]. 地球科学: 中国地质大学学报, 2009, 34(6): 921-930.
[37]	JIANG W C, LI H, EVANS N J, et al. Zircon records multiple magmatic-hydrothermal processes at the giant Shizhuyuan W-Sn-Mo-Bi polymetallic deposit, South China[J]. Ore Geology Reviews, 2019, 115: 103160.
[38]	TOSCANO M, PASCUAL E, NESBITT R W, et al. Geochemical discrimination of hydrothermal and igneous zircon in the Iberian pyrite belt, Spain[J]. Ore Geology Reviews, 2014, 56: 301-311. DOI URL
[39]	SUN Y, WU T, XIAO L, et al. U-Pb ages, Hf-O isotopes and trace elements of zircons from the ore-bearing and ore-barren adakitic rocks in the Handan-Xingtai district: implications for petrogenesis and iron mineralization[J]. Ore Geology Reviews, 2019, 104: 14-25. DOI URL
[40]	WANG M, GUO W F, YANG W T. Detrital zircon trace elements from the Mesozoic Jiyuan Basin, central China and its implication on tectonic transition of the Qinling Orogenic Belt[J]. Open Geosciences, 2019, 11(1): 125-139. DOI URL
[41]	WANG R, JEON H, EVANS N J. Archaean hydrothermal fluid modified zircons at Sunrise Dam and Kanowna Belle gold deposits, Western Australia: implications for post-magmatic fluid activity and ore genesis[J]. American Mineralogist, 2018, 103(12): 1891-1905. DOI URL
[42]	ZENG L J, NIU H C, BAO Z W, et al. Chemical lattice expansion of natural zircon during the magmatic-hydrothermal evolution of A-type granite[J]. American Mineralogist, 2017, 102(3): 655-665. DOI URL
[43]	ZHANG W, JIANG S Y, GAO T S, et al. The effect of magma differentiation and degassing on ore metal enrichment during the formation of the world-class Zhuxi W-Cu skarn deposit: evidence from U-Pb ages, Hf isotopes and trace elements of zircon, and whole-rock geochemistry[J]. Ore Geology Reviews, 2020, 127: 103801. DOI URL
[44]	WU M Q, TIAN B F, ZHANG D H, et al. Zircon of the No. 782 deposit from the Great Xing'an Range in NE China: implications for Nb-REE-Zr mineralization during magmatic-hydrothermal evolution[J]. Ore Geology Reviews, 2018, 102: 284-299. DOI URL
[45]	ZHANG S H, ZHAO Y, YANG Z Y, et al. The 1.35 Ga diabase sills from the northern North China Craton: implications for breakup of the Columbia (Nuna) supercontinent[J]. Earth and Planetary Science Letters, 2009, 288(3/4): 588-600. DOI URL
[46]	ZHAO L, GUO F, FAN W M, et al. Early Cretaceous potassic volcanic rocks in the Jiangnan Orogenic Belt, East China: crustal melting in response to subduction of the Pacific-Izanagi ridge?[J]. Chemical Geology, 2016, 437: 30-43. DOI URL
[47]	ZHU M T, ZHANG L C, DAI Y P, et al. Hydrothermal modification of zircon geochemistry and Lu-Hf isotopes from the Hongtoushan Cu-Zn deposit, China[J]. Ore Geology Reviews, 2017, 86: 707-718. DOI URL
[48]	秦亚, 冯佐海, 黄靖哲, 等. 桂北地区三门韧性剪切带的厘定及其构造意义[J]. 地球科学, 2021, 46(11): 4017-4032.
[49]	LI H, LI J W, ALGEO T J, et al. Zircon indicators of fluid sources and ore genesis in a multi-stage hydrothermal system: the Dongping Au deposit in North China[J]. Lithos, 2018, 314/315: 463-478. DOI URL
[50]	TAKEHARA M, HORIE K, HOKADA T, et al. New insight into disturbance of U-Pb and trace-element systems in hydrothermally altered zircon via SHRIMP analyses of zircon from the Duluth Gabbro[J]. Chemical Geology, 2018, 484: 168-178. DOI URL
[51]	LEI W Y, SHI G H, SANTOSH M, et al. Trace element features of hydrothermal and inherited igneous zircon grains in mantle wedge environment: a case study from the Myanmar jadeitite[J]. Lithos, 2016, 266/267: 16-27. DOI URL
[52]	BAO Z W, SUN W D, LI C J, et al. U-Pb dating of hydrothermal zircon from the Dongping gold deposit in North China: constraints on the mineralization processes[J]. Ore Geology Reviews, 2014, 61: 107-119. DOI URL
[53]	WU J H, LI H, ALGEO T J, et al. Genesis of the Xianghualing Sn-Pb-Zn deposit, South China: a multi-method zircon study[J]. Ore Geology Reviews, 2018, 102: 220-239. DOI URL
[54]	HAN J S, CHEN H Y, HOLLINGS P, et al. The formation of modified zircons in F-rich highly-evolved granites: an example from the Shuangji granites in Eastern Tianshan, China[J]. Lithos, 2019, 324/325: 776-788. DOI URL
[55]	JIANG W C, LI H, TURNER S, et al. Timing and origin of multi-stage magmatism and related W-Mo-Pb-Zn-Fe-Cu mineralization in the Huangshaping deposit, South China: an integrated zircon study[J]. Chemical Geology, 2020, 552: 119782. DOI URL
[56]	WEBER B, SCHMITT A K, CISNEROS DE LEÓN A, et al. Neoproterozoic extension and the Central Iapetus Magmatic Province in southern Mexico- New U-Pb ages, Hf-O isotopes and trace element data of zircon from the Chiapas Massif Complex[J]. Gondwana Research, 2020, 88: 1-20. DOI URL
[57]	YANG Y, LIANG C Y, ZHENG C Q, et al. Metamorphic evolution of high-grade granulite-facies rocks of the Mashan Complex, Liumao area, eastern Heilongjiang Province, China: evidence from zircon U-Pb geochronology, geochemistry and phase equilibria modelling[J]. Precambrian Research, 2021, 355: 106095. DOI URL
[58]	KIRKLAND C L, SLAGSTAD T, JOHNSON T E. Zircon as a metamorphic timekeeper: a case study from the Caledonides of central Norway[J]. Gondwana Research, 2018, 61: 63-72. DOI URL
[59]	ZHANG S B, TANG J, ZHENG Y F. Contrasting Lu-Hf isotopes in zircon from Precambrian metamorphic rocks in the Jiaodong Peninsula: constraints on the tectonic suture between North China and South China[J]. Precambrian Research, 2014, 245: 29-50. DOI URL
[60]	XIE S W, WU Y B, ZHANG Z M, et al. U-Pb ages and trace elements of detrital zircons from Early Cretaceous sedimentary rocks in the Jiaolai Basin, north margin of the Sulu UHP terrane: provenances and tectonic implications[J]. Lithos, 2012, 154: 346-360. DOI URL
[61]	KOOIJMAN E, UPADHYAY D, MEZGER K, et al. Response of the U-Pb chronometer and trace elements in zircon to ultrahigh-temperature metamorphism: the Kadavur anorthosite complex, southern India[J]. Chemical Geology, 2011, 290(3/4): 177-188. DOI URL
[62]	ROCHA B C, MORAES R, MÖLLER A, et al. Timing of anatexis and melt crystallization in the Socorro-Guaxupé Nappe, SE Brazil: insights from trace element composition of zircon, monazite and garnet coupled to U-Pb geochronology[J]. Lithos, 2017, 277: 337-355. DOI URL
[63]	ZHONG R C, DENG Y, YU C. Multi-layer perceptron-based tectonic discrimination of basaltic rocks and an application on the Paleoproterozoic Xiong'er volcanic province in the North China Craton[J]. Computers and Geosciences, 2021, 149: 104717. DOI URL
[64]	HELSEL D R. More than obvious: better methods for interpreting nondetect data[J]. Environmental Science and Technology, 2005, 39(20): 419A-423A.
[65]	ROUSSEEUW P J. Silhouettes: a graphical aid to the interpretation and validation of cluster analysis[J]. Journal of Computational and Applied Mathematics, 1987, 20: 53-65. DOI URL
[66]	CHAWLA N V, BOWYER K W, HALL L O, et al. SMOTE: synthetic minority over-sampling technique[J]. Journal of Artificial Intelligence Research, 2002, 16: 321-357. DOI URL
[67]	CRACKNELL M J, READING A M. Geological mapping using remote sensing data: a comparison of five machine learning algorithms, their response to variations in the spatial distribution of training data and the use of explicit spatial information[J]. Computers and Geosciences, 2014, 63: 22-33. DOI URL
[68]	BREIMAN L. Arcing classifiers[J]. The Annals of Statistics, 1998, 26(3): 801-824.
[69]	BREIMAN L. Using iterated bagging to debias regressions[J]. Machine Learning, 2001, 45(3): 261-277. DOI URL
[70]	韩帅, 李明超, 任秋兵, 等. 基于大数据方法的玄武岩大地构造环境智能挖掘判别与分析[J]. 岩石学报, 2018, 34(11): 3207-3216.
[71]	CHANG C C, LIN C J. Libsvm[J]. ACM Transactions on Intelligent Systems and Technology, 2011, 2(3): 1-27.
[72]	HSU C W, CHANG C C, LIN C J. A practical guide to support vector classification[J]. BJU International, 2008, 101(1): 1396-1400. DOI URL
[73]	KOHONEN T. An introduction to neural computing[J]. Neural Networks, 1988, 1(1): 3-16.
[74]	BELOUSOVA E, GRIFFIN W, O'REILLY S Y, et al. Igneous zircon: trace element composition as an indicator of source rock type[J]. Contributions to Mineralogy and Petrology, 2002, 143(5): 602-622. DOI URL
[75]	邹心宇, 蒋济莲, 秦克章, 等. 锆石微量元素的理论基础及其应用研究进展[J]. 岩石学报, 2021, 37(4): 985-999.

基于机器学习的锆石成因分类研究

Machine learning-based approach for zircon classification and genesis determination

RichHTML

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

图/表 12

参考文献 75

相关文章 15

编辑推荐

Metrics

本文评价

[1]	吴浩, 杨晨, 吴彦旺, 李才, 刘飞, 林兆旭. 藏北中仓地区晚白垩世岩浆岩成因及其对高原早期隆升的指示[J]. 地学前缘, 2024, 31(6): 261-281.
[2]	吴庆, 黄芬, 郭永丽, 肖琼, 孙平安, 杨慧, 白冰. 西南岩溶小流域水体中微量元素地球化学特征及其指示意义[J]. 地学前缘, 2024, 31(5): 397-408.
[3]	刘伟, 张洪瑞, 罗迪柯, 贾鹏飞, 靳立杰, 周永刚, 梁云汉, 王子圣, 李春稼. 安哥拉地块北部Dondo地区古元古代花岗岩岩石成因:Columbia超大陆聚合的响应[J]. 地学前缘, 2024, 31(4): 237-257.
[4]	张焕宝, 贺海洋, 杨仕教, 李亚林, 毕文军, 韩世礼, 郭钦鹏, 杜青. 基于机器学习的埃达克质岩构造背景判别研究[J]. 地学前缘, 2024, 31(4): 417-428.
[5]	陈国超, 张晓飞, 裴先治, 裴磊, 李佐臣, 刘成军, 李瑞保. 雅鲁藏布江中段日喀则地区却顶布—路曲地幔橄榄岩岩石地球化学特征、成因及其地质意义[J]. 地学前缘, 2024, 31(3): 1-19.
[6]	尹青青, 唐菊兴, 项新葵, 赵晓彦, 汪方跃, 徐裕敏, 郭虎, 余振东, 谢金玲, 代晶晶, 彭勃. 赣北彭山还原性S型花岗岩成因及其对Sn富集的启示:来自锆石微量元素的证据[J]. 地学前缘, 2024, 31(3): 133-149.
[7]	陈可, 邵拥军, 刘忠法, 张俊柯, 李永顺, 陈雨莹. 岩浆因素对中国东部铜陵矿集区差异性矿化的控制作用:来自角闪石、斜长石矿物学证据[J]. 地学前缘, 2024, 31(3): 199-217.
[8]	张家志, 姜在兴, 徐杰, 魏思源, 宋立舟, 刘桐, 沈志晗, 姜晓龙, 李永飞, 张玺. 朝阳盆地白垩系九佛堂组火山沉积作用及其对有机质富集的影响[J]. 地学前缘, 2024, 31(3): 284-297.
[9]	张健, 何雨蓓, 范艳霞. 福建沿海地区地热异常热源成因的地球物理分析[J]. 地学前缘, 2024, 31(3): 392-401.
[10]	张利军, 鲁文豪, 张建东, 彭光雄, 卜建财, 唐凯, 谢渐成, 徐质彬, 杨海燕. 基于深度学习的镜下岩石、矿物薄片识别[J]. 地学前缘, 2024, 31(3): 498-510.
[11]	刘洋, 李三忠, 钟世华, 郭广慧, 刘嘉情, 牛警徽, 薛梓萌, 周建平, 董昊, 索艳慧. 机器学习:海底矿产资源智能勘探的新途径[J]. 地学前缘, 2024, 31(3): 520-529.
[12]	苏恺明, 徐耀辉, 徐旺林, 张月巧, 白斌, 李阳, 严刚. 鄂尔多斯盆地延长组多油源贡献比例与分布规律:基于机器学习与可解释性研究[J]. 地学前缘, 2024, 31(3): 530-540.
[13]	刘持恒, 李子颖, 贺锋, 张字龙, 李振成, 凌明星, 刘瑞萍. 鄂尔多斯盆地西北部下白垩统物源定量分析研究[J]. 地学前缘, 2024, 31(3): 80-99.
[14]	钏茂山, 胡乐, 蔺如喜, 毛崇祯, 李仕忠, 李锁明, 袁永盛. 扬子板块西缘早中生代“绿豆岩”成因及构造启示:锆石U-Pb年龄、微量元素及Hf同位素约束[J]. 地学前缘, 2024, 31(2): 204-223.
[15]	周予茜, 时毓, 黄椿文, 刘希军, 蓝媛春, 唐源远, 翁伯寅. 桂东南莲垌和古龙岩体加里东期I型花岗岩类的岩石成因及构造意义[J]. 地学前缘, 2024, 31(2): 224-248.