Machine-readable expression of unstructured geological information and intelligent prediction of mineralization associated anomaly areas in Pangxidong District, Guangdong, China

doi:10.13745/j.esf.sf.2024.5.5

Abstract

Abstract:

The application of big data mining and machine learning algorithms in mineralization prediction has become an important research trend, but unstructured geological data cannot be directly mined—first they need to be converted to machine-readable expressions. In this study of the Pangxidong ore district in western Guangdong Province, the unstructured geological information such as stratigraphy, lithology, faults are converted into machine-readable format, and two machine learning algorithms, namely, One-Class Support Vector Machine and Auto-Encoder Network, are applied to mine the geochemical test data of stream sediments as well as the comprehensive geological information on faults, stratigraphy, etc. to extract the features of mineralization anomalies and ultimately achieve intelligent delineation of the anomaly areas. Through combined application of One-Hot Encoder and the weighted variable method for spatially weighted principal component analysis, the structural transformation of the unstructured geological information is realized, and geological information is maximally preserved for data mining. It is demonstrated that the application of One-Class Support Vector Machine and Auto-Encoder Network can effectively solve the problem of data imbalance, as the numbers of ore and non-ore spots in the study area are seriously unbalanced. The prediction results generated using the integrated, synthesized multi-source geological data are relatively consistent with the observed spatial distribution of Pb-Zn deposits and the actual geological structure in the study area, indicating the two algorithms can effectively identify potential prospecting targets and ore deposits. Compared with traditional geochemical prospecting methods, the intelligent prediction method can process and integrate multi-source geological information about the ore-forming processes and identify mineralization anomaly areas. This method is applicable in prospecting areas without prior ore discovery, thereby improving the efficiency of ore prospecting and increasing the possibility of finding ore deposits.

Key words: big data mining, machine-readable expression, One-Hot Encoder, One-Class Support Vector Machine, Auto-Encoder Network, Pangxidong ore district, Qinzhou-Hangzhou metallogenic belt

CLC Number:

WANG Kunyi, ZHOU Yongzhang. Machine-readable expression of unstructured geological information and intelligent prediction of mineralization associated anomaly areas in Pangxidong District, Guangdong, China[J]. Earth Science Frontiers, 2024, 31(4): 47-57.

Figures/Tables 10

Fig.1 Simplified regional geological map of the Pangxidong ore district. Modified after [8].

Fig.2 Schematic diagram of Support Vector Data Description (SVDD). Modified after [17].

Fig.3 Structural diagram of Auto-Encoder Network. Modified after [22].

Table 1 Characteristics of machine-readable expression data for stratigraphy, lithology, and faults (partial results)

X	Y	塘蓬群	帽子峰组	天子岭组	东岗岭组	信都组	老虎头组	新元古代混合岩	燕山早期侵入岩	燕山晚期侵入岩	空间权重变量w
442 920	2 410 830	0	0	0	0	1	0	0	0	0	0.83
434 920	2 424 830	1	0	0	0	0	0	0	0	0	0.79
439 170	2 408 580	0	0	0	0	0	1	0	0	0	0.77
430 670	2 433 830	0	0	0	0	0	0	0	1	0	0.88
418 170	2 400 080	0	1	0	0	0	0	0	0	0	0.86
402 670	2 425 580	0	0	0	0	0	0	0	0	1	0.92
431 670	2 398 580	0	0	0	1	0	0	0	0	0	0.93
424 670	2 399 080	0	0	1	0	0	0	0	0	0	0.94
401 670	2 425 830	0	0	0	0	0	0	0	0	1	0.96
400 170	2 434 080	0	0	0	0	0	0	1	0	0	0.98
445 170	2 404 080	0	1	0	0	0	0	0	0	0	0.99

Fig.4 Correspondence relationship between t-value for known Pb-Zn deposits in the study area and buffer zone distance of NW-trending faults

Fig.5 Weight coefficient map from spatially weighted principal component analysis of NW-trending faults in the Pangxidong area

Fig.6 Results of cluster analysis of geochemical IDW interpolated data for 16 elements

Fig.7 Mineralization anomaly map by One-Class Support Vector Machine

Fig.8 Mineralization anomaly map by Auto-Encoder Network

Fig.9 Receiver operating characteristic curves obtained by One-Class Support Vector Machine (a) and Auto-Encoder Network (b)

References 34

[1]	周永章, 张良均, 张奥多, 等. 地球科学大数据挖掘与机器学习[M]. 广州: 中山大学出版社, 2018.
[2]	翟明国, 杨树锋, 陈宁华, 等. 大数据时代: 地质学的挑战与机遇[J]. 中国科学院院刊, 2018, 33(8): 825-831.
[3]	成秋明. 什么是数学地球科学及其前沿领域?[J]. 地学前缘, 2021, 28(3): 6-25. DOI
[4]	左仁广. 勘查地球化学数据挖掘与弱异常识别[J]. 地学前缘, 2019, 26(4): 67-75. DOI
[5]	刘艳鹏, 朱立新, 周永章. 卷积神经网络及其在矿床找矿预测中的应用: 以安徽省兆吉口铅锌矿床为例[J]. 岩石学报, 2018, 34(11): 3217-3224.
[6]	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.
[7]	周永章, 李兴远, 郑义, 等. 钦杭结合带成矿地质背景及成矿规律[J]. 岩石学报, 2017, 33(3): 667-681.
[8]	周永章, 张国桓, 吴勇庆, 等. 广东庞西垌地区矿产远景调查报告(文地幅、石角幅、塘蓬幅、河唇幅, 1∶50000)[R]. 北京: 中国地质调查局, 2016.
[9]	周永章, 曾长育, 李红中, 等. 钦州湾-杭州湾构造结合带(南段)地质演化和找矿方向[J]. 地质通报, 2012, 31(2/3): 486-491.
[10]	广东省地质矿产局704地质大队. 中华人民共和国区域地质调查报告1∶50000塘蓬幅[R]. 湛江: 广东省地质矿产局 704地质大队, 1987.
[11]	广东省地质矿产局704地质大队. 中华人民共和国区域地质调查报告1∶50000河唇幅[R]. 湛江: 广东省地质矿产局 704地质大队, 1994.
[12]	广东省地质矿产局. 广东省区域地质志[R]. 北京: 地质出版社, 1988, 941.
[13]	战明国, 彭松柏, 蔡明海, 等.云开地区重要成矿区带金、银、铜、铅、锌成矿地质背景及找矿靶区优选研究[M]. 海口: 海南出版社, 2006.
[14]	DAVY M, GODSILL S. Detection of abrupt spectral changes using support vector machines: an application to audio signal segmentation[C]// Proceedings of IEEE international conference on acoustics speech and signal processing. Orlando: IEEE, 2002: 1313-1316.
[15]	LECOMTE S, LENGELLE R, RICHARD C, et al. Abnormal events detection using unsupervised One-Class SVM - Application to audio surveillance and evaluation[C]// Proceedings of 2011 8th IEEE international conference on advanced video and signal based surveillance (AVSS). Klagenfurt: IEEE, 2011: 124-129.
[16]	SCHÖLKOPF B, PLATT J C, SHAWE-TAYLOR J C, et al. Estimating the support of a high-dimensional distribution[J]. Neural Computation, 2001, 13(7): 1443-1471. PMID
[17]	吴定海, 张培林, 任国全, 等. 基于支持向量的单类分类方法综述[J]. 计算机工程, 2011, 37(5): 187-189. DOI
[18]	TAX D M J, DUIN R P W. Support vector data description[J]. Machine Learning, 2004, 54(1): 45-66.
[19]	VAPNIK V N. The nature of statistical learning theory[M]. New York: Springer, 1995.
[20]	RUMELHART D E, HINTON G E, WILLIAMS R J. Learning representations by back-propagating errors[J]. Nature, 1986, 323: 533-536.
[21]	AN J, CHO S. Variational autoencoder based anomaly detection using reconstruction probability[J]. Special Lecture on IE, 2015, 2(1): 1-18.
[22]	VALENTINE A P, TRAMPERT J. Data space reduction, quality assessment and searching of seismograms: autoencoder networks for waveform data[J]. Geophysical Journal International, 2012, 189(2): 1183-1202.
[23]	RIFAI S, VINCENT P, MULLER X, et al. Contractive auto-encoders: explicit invariance during feature extraction[C]// Proceedings of the 28th international conference on machine learning. Bellevue: ACM, 2011: 833-840.
[24]	吴冲龙, 刘刚, 张夏林, 等. 地质科学大数据及其利用的若干问题探讨[J]. 科学通报, 2016, 61(16): 1797-1807.
[25]	王成彬, 马小刚, 陈建国. 数据预处理技术在地学大数据中应用[J]. 岩石学报, 2018, 34(2): 303-313.
[26]	张雪英, 叶鹏, 王曙, 等. 基于深度信念网络的地质实体识别方法[J]. 岩石学报, 2018, 34(2): 343-351.
[27]	张雪英, 张春菊, 汪陈, 等. 面向中文文本的地质语义信息标注与语料库构建[J]. 高校地质学报, 2023, 29(3): 429-438.
[28]	CHENG Q M, BONHAM-CARTER G, WANG W L, et al. A spatially weighted principal component analysis for multi-element geochemical data for mapping locations of felsic intrusions in the Gejiu mineral district of Yunnan, China[J]. Computers and Geosciences, 2011, 37(5): 662-669.
[29]	XIAO F, CHEN J G, ZHANG Z Y, et al. Singularity mapping and spatially weighted principal component analysis to identify geochemical anomalies associated with Ag and Pb-Zn polymetallic mineralization in Northwest Zhejiang, China[J]. Journal of Geochemical Exploration, 2012, 122: 90-100.
[30]	肖凡, 陈建国, 侯卫生, 等. 钦-杭结合带南段庞西垌地区Ag-Au致矿地球化学异常信息识别与提取[J]. 岩石学报, 2017, 33(3): 779-790.
[31]	AGTERBERG F P, BONHAM-CARTER G F, WRIGHT D F. Statistical pattern integration for mineral exploration[M]// Computer applications in resource estimation. Amsterdam: Elsevier, 1990: 1-21.
[32]	BONHAM-CARTER G F. Geographic information systems for geoscientists: modelling with GIS[M]. Amsterdam: Elsevier, 1994.
[33]	YU X T, XIAO F, ZHOU Y Z, et al. Application of hierarchical clustering, singularity mapping, and Kohonen neural network to identify Ag-Au-Pb-Zn polymetallic mineralization associated geochemical anomaly in Pangxidong district[J]. Journal of Geochemical Exploration, 2019, 203: 87-95.
[34]	余晓彤, 肖凡, 周永章, 等. 粤西庞西垌地区银金地球化学异常信息挖掘与提取[J]. 地质与勘探, 2019, 55(1): 77-86.