Earth Science Informatics

, Volume 11, Issue 4, pp 553–566 | Cite as

Back-propagation neural network and support vector machines for gold mineral prospectivity mapping in the Hatu region, Xinjiang, China

  • Nannan ZhangEmail author
  • Kefa Zhou
  • Dong Li
Research Article


Machine Learning technologies have the potential to deliver new nonlinear mineral prospectivity mapping (MPM) models. In this study, Back Propagation (BP) neural network Support Vector Machine (SVM) methods were applied to MPM in the Hatu region of Xinjiang, northwestern China. First, a conceptual model of mineral prospectivity for Au deposits was constructed by analysis of geological background. Evidential layers were selected and transformed into a binary data format. Then, the processes of selecting samples and parameters were described. For the BP model, the parameters of the network were 9–10 − 1; for the SVM model, a radial basis function was selected as the kernel function with best C = 1 and γ = 0.25. MPM models using these parameters were constructed, and threshold values of prediction results were determined by the concentration-area (C-A) method. Finally, prediction results from the BP neural network and SVM model were compared with that of a conventional method that is the weight- of- evidence (W- of- E). The prospectivity efficacy was evaluated by traditional statistical analysis, prediction-area (P-A) plots, and the receiver operating characteristic (ROC) technique. Given the higher intersection position (74% of the known deposits were within 26% of the total area) and the larger AUC values (0.825), the result shows that the model built by the BP neural network algorithm has a relatively better prediction capability for MPM. The BP neural network algorithm applied in MPM can elucidate the next investigative steps in the study area.


Mineral prospectivity mapping Back propagation neural network Support vector machines Weight of evidence 



This work was supported by the National Natural Science Foundation of China (Grant No. 41602339, U1503291), the Western Light Foundation of the Chinese Academy of Sciences (CAS; Grant No. XBBS-2014-19), the Xinjiang Uygur Autonomous Major Project (Grant No. 201330121-3), the National Basic Research Program of China (Grant No. 973Program2014CB440803) and the Strategic Priority Research Program of the Chinese Academy of Sciences(XDA19030204).


  1. Abedi M, Norouzi GH (2012) Integration of various geophysical data with geological and geochemical data to determine additional drilling for copperexploration. J Appl Geophys 83:35–45CrossRefGoogle Scholar
  2. Abedi M, Torabi SA, Norouzi GH, Hamzeh M, Elyasi GR (2012a) PROMETHEEII: a knowledge-driven method for copper exploration. Comput Geosci 46:255–263CrossRefGoogle Scholar
  3. Abedi M, Torabi SA, Norouzi GH, Hamzeh M (2012b) ELECTRE III: a knowledgedriven method for integration of geophysical data with geological and geochemical data in mineral prospectivity mapping. J Appl Geophys 87:9–18CrossRefGoogle Scholar
  4. Abedi M, Norouzi GH, Fathianpour N (2013) Fuzzy outranking approach: a knowledge-driven method for mineral prospectivity mapping. Int J Appl Earth Obs Geoinf 21:556–567CrossRefGoogle Scholar
  5. Abedi M, Kashani SBM, Norouzi GH, Yousefi M (2017) A deposit scale mineral prospectivity analysis: a comparison of various knowledge-driven approaches for porphyry copper targeting in Seridune, Iran. J Afr Earth Sci 128:127–146CrossRefGoogle Scholar
  6. Agterberg FP (1971) A in the forward propagation process, index for detecting favourable geological environments. Can Inst Min Metall 10:82–91Google Scholar
  7. Agterberg FP (1974) Automatic contouring of geological maps to detect target areas for mineral exploration. Math Geol 6:373–395CrossRefGoogle Scholar
  8. Agterberg FP, Bonham-Carter GF (1999) Logistic regression and weights of evidence modeling in mineral exploration, Proc.28th Int Symp app Comput mineral Ind (APCOM). Golden,CO, USA, pp 483–490Google Scholar
  9. Atkinson P, Tatnall A (1997) Introduction neural networks in remote sensing. IntJ Remote Sens 18:699–709CrossRefGoogle Scholar
  10. Bonham-Carter GF (1994) Geographic information Systems for Geoscientists: modeling with GIS. Pergamon Press, Ontario, Canada 398Google Scholar
  11. Bonham-Carter GF, Agterberg FP, Wright DF (1989) Weights of evidence modelling: a new approach to mapping mineralpotential, statistical applications in earth. Sciences 89(9):171–183Google Scholar
  12. Bradley AP (1997) The use of the area under the ROC curve in the evaluation of machine learning algorithms. Pattern Recogn 30:1145–1159CrossRefGoogle Scholar
  13. Breiman L (2001) Randomforests. Mach. Learn 45:5–32CrossRefGoogle Scholar
  14. Brown WM, Gedeon TD, Groves DI, Barnes RG (2000) Artificial neural networks: a new method for mineral prospectivity mapping. Aust J Earth Sci 47:757–770CrossRefGoogle Scholar
  15. Burger H, Kirsch C, Skala W (1989). The application of microcomputers in exploration and exploitation of mineral deposits.Original Research Article Computers & Geosciences 15(4): 587–591CrossRefGoogle Scholar
  16. Carranza EJM (2008) Geochemical anomaly and mineral prospectivity mapping in GIS. In: Handbook of exploration and environmental geochemistry, vol 11. Elsevier, Amsterdam, p 351Google Scholar
  17. Carranza EJM (2010) Improved wildcat modelling of mineral prospectivity. Resour Geol 60:129–149CrossRefGoogle Scholar
  18. Carranza EJM (2017) Natural resources research publications on geochemical anomaly and mineral potential mapping, and introduction to the special issue of papers in these fields. Nat Resour Res 26(4):379–410CrossRefGoogle Scholar
  19. Carranza EJM, Hale M (2001) Logistic regression for geologically constrained mapping of gold potential, Baguio district. Philipp. Explor Min Geol 10:165–175CrossRefGoogle Scholar
  20. Carranza EJM, Hale M (2002a) Wildcat mapping of gold potential, Baguio district,Philippines. Trans Inst Min Metall Appl Earth Sci 111:100–105CrossRefGoogle Scholar
  21. Carranza EJM, Hale M (2002b) Where porphyry copper deposits are spatially localized? A case study in Benguet province, Philippines. Nat Resour Res 11:45–59CrossRefGoogle Scholar
  22. Carranza EJM, Mangaoang JC, Hale M (1999) Application of mineral exploration models and GIS to generate mineral potential maps as input for optimum landuse planning in the Philippines. Nat Resour Res 8:165–173CrossRefGoogle Scholar
  23. Chen Y (2015) Mineral potential mapping with a restricted Boltzmann machine. Ore Geol Rev 71:749–760CrossRefGoogle Scholar
  24. Chen Y, Lu L, Li X (2014) Application of continuous restricted Boltzmann machine to identify multivariate geochemical anomaly. J Geochem Explor 140:56–63CrossRefGoogle Scholar
  25. Cheng Y, Wu W (2017) Mapping mineral prospectivity using an extreme learning machine regression. Ore Geol Rev 80:200–213CrossRefGoogle Scholar
  26. Cheng Q, Agterberg FP, Ballantyne SB (1994) The separation of geochemical anomalies from background by fractal methods. J Geochem Explor 51:109–130CrossRefGoogle Scholar
  27. Cheng Q, Chen ZJ, Khaled A (2007) Application of fuzzyweights of evidence methodin mineral resource assessmentfor gold in Zhenyuan District, Yunnan Province, China. Earth Sci J China Univ Geosci (In Chinese) 32:175–184Google Scholar
  28. David BS, Paul KT, Shen SQ et al (1993) The Hatu gold anomaly, Xinjiang-Uygur autonomous region, China — testing the hypothesis of aeolian transport of gold. J Geochem Explor 47:201–216CrossRefGoogle Scholar
  29. Fawcett T (2006) An introduction to ROC analysis. Pattern Recogn Lett 27:861–874CrossRefGoogle Scholar
  30. Foody GM, Mathur A (2004) A relative evaluation of multiclass image classification by support vector machines. IEEE Trans Geosci Remote Sens 42(6):1335–1343CrossRefGoogle Scholar
  31. Gao Y, Zhang ZJ, Xiong YH, Zuo RG (2016) Mapping mineral prospectivity for cu polymetallic mineralization insouthwest Fujian Province, China .Ore Geol Rev 75 (2016) 16–28CrossRefGoogle Scholar
  32. Good IJ (1950) Probability and the weighting of evidence. Griffin, London 119ppGoogle Scholar
  33. Hamid NA, Nawi NM, Ghazali R (2011) Accelerating Learning Performance of Back Propagation Algorithm by Using Adaptive Gain Together with Adaptive Momentum and Adaptive Learning Rate on Classification Problems Computing and Multimedia Applications. Second International Conference, UCMA 2011, Daejeon, Korea, April 13–15. Proceedings, Part IIGoogle Scholar
  34. Han BF, Qinq JJ, Sun B et al (2017) Late Paleozoic vertical growth of continental crust around the Junggar Basin, Xinjiang, China (Part I) : Timing of post-collisional plutonism. Acta Petrol Sin 2006 22(5):1077–1086 (in Chinese)Google Scholar
  35. Harris DP (1965) An Application of Multivariate Statistical Analysis to Mineral ExplorationGoogle Scholar
  36. Harris DP (1969) Alaska’s base and precious metals resources: a probabilistic regional appraisal. Q. J. Colorado Sch. Min 64:295–327Google Scholar
  37. Harris D, Pan G (1999) Mineral favorability mapping: a comparison of artificial neural networks, logistic regression, and discriminant analysis. Nat Resour Res 8:93–109CrossRefGoogle Scholar
  38. Hashemi Tangestani M, Moore F (2002) The use of Dempster-Shafer model andGIS in integration of geoscientific data for porphyry copper potential mapping,north of Shahr-e-Babak, Iran. Int J Appl Earth Observ Geoinform 4:65–74CrossRefGoogle Scholar
  39. Liu Y, Zhou KF, Xia QL (2017) A MaxEnt model for mineral Prospectivity mapping. Nat Resour Res. CrossRefGoogle Scholar
  40. Ma D, Zhou T, Chen J et al (2017) Supercritical water heat transfer coefficient prediction analysis based on BP neural network. Nucl Eng Des 320:400–408CrossRefGoogle Scholar
  41. Mejía-Herrera P, Royer JJ, Caumon G, Cheilletz A (2015) Curvature attribute from surface-restoration as predictor variable in Kupferschiefer copper potentials: an example from the fore-Sudetic region. Nat Resour Res 24(3):275–290CrossRefGoogle Scholar
  42. Molan YE, Behnia P (2013) Prospectivity mapping of Pb–Zn SEDEXmineralization using remote-sensing data inthe Behabad area, Central Iran. Int J Remote Sens 34(4):1164–1179CrossRefGoogle Scholar
  43. Moon WM (1990) Integration of geophysical and geological data using evidential belief function. IEEE Trans Geosci Remote Sens 28:711–720CrossRefGoogle Scholar
  44. Najafi A, Karimpour MH, Ghaderi M (2014) Application of fuzzy AHP method to IOCG prospectivity mapping Acase study in Taherabad prospecting area, eastern Iran. Int J Appl Earth Obs Geoinf 33:142–154CrossRefGoogle Scholar
  45. Nykänen V, Lahti I, Niiranen T, Korhonen K (2015) Receiver operating characteristics (ROC) as validation tool for prospectivity models — a magmatic Ni–cu case study from the Central Lapland Greenstone Belt, northern Finland. Ore Geol Rev 71:853–860CrossRefGoogle Scholar
  46. Oh H, Lee S (2010) Application of artificial neural network for gold-silver deposits potential mapping: a case study of Korea. Nat Resour Res 19:103–124CrossRefGoogle Scholar
  47. Oommen T, Misra D, Twarakavi NKC et al (2008) An objective analysis of support vector machine based classification for remote sensing. Math Geosci 40:409–422CrossRefGoogle Scholar
  48. Piccini C, Marchetti A, Farina R, Francaviglia R (2012) Application of indicator kriging to evaluate the probability of exceeding nitrate contamination thresholds. Int J Environ Res 6(4):853–862Google Scholar
  49. Porwal A, Carranza EJM, Hale M (2006) Bayesian network classifiers formineral potential mapping. Comput Geosci 32(1):1–16CrossRefGoogle Scholar
  50. Rodriguez-Galiano V, Sanchez-Castillo M, Chica-Olmo M et al (2014) Machine learning predictive models for mineral prospectivity: Anevaluation of neural networks, random forest, regression trees and support vector machines. Ore Geol Rev 71:804–818CrossRefGoogle Scholar
  51. Sadeghi B, Khalajmasoumi M (2015) A futuristic review for evaluation of geothermal potentials using fuzzy logic and binary index overlay in GIS environment. Renew Sust Energ Rev 43:818–831CrossRefGoogle Scholar
  52. Shen P, Shen YC, Liu TB et al (2009) Geochemical signature of porphyries in the Baogutu porphyry copper belt, western Junggar, NW China. Gondwana Res 16(2):227–242CrossRefGoogle Scholar
  53. Shen P, Pan HD, Zhu HP (2016) Two fluid sources and genetic implications for the Hatu gold deposit, Xinjiang, China. Ore Geol Rev 73(2):298–312CrossRefGoogle Scholar
  54. Sinclair AJ, Woodsworth GL (1970) Multiple regression as a method of estimating exploration potential in an area near terrace, B.C. Econ Geol 65:998–1003CrossRefGoogle Scholar
  55. Smola AJ, Schölkopf B (2004) A tutorial on support vector regression. Stat Comput 14(3):199–222CrossRefGoogle Scholar
  56. Wang L, Zhu YF (2015) Multi-stage pyrite and hydrothermal mineral assemblage of the Hatu gold district (west Junggar, Xinjiang, NW China): implications for metallogenic evolution. Ore Geol Rev 69:243–267CrossRefGoogle Scholar
  57. Xu YJ, You T, Cl D (2015) An integrated micromechanical model and BP neural network for predicting elastic modulus of 3-D multi-phase and multi-layer braided composite. Compos Struct 122:308–315CrossRefGoogle Scholar
  58. Yin LB, Liu GC, Zhou JL et al (2017) A calculation method for CO2 emission in utility boilers based on BP neural network and carbon balance. Energy Procedia 105:3173–3178CrossRefGoogle Scholar
  59. Yousefi M, Kamkar-Rouhani A, Carranza EJM (2012) Geochemical mineralization probability index(GMPI):a new approach to generateen hanced stream sediment geochemical evidential map for increasing probability of success in mineral potential mapping. J Geochem Explor 115:24–35CrossRefGoogle Scholar
  60. Yousefi M, Carranza EJM, Kamkar-Rouhani A (2013) Weighteddrainagecatchment basin mapping of stream sediment geochemical anomalies for mineral potential mapping. J Geochem Explor 128:88–96CrossRefGoogle Scholar
  61. Yule GU (1912) On the methods of measuring association between two attributes. J.R. Stat. Soc 75:579–642CrossRefGoogle Scholar
  62. Zhang NN, Zhou KF (2015) Mineral prospectivity mapping with weights of evidence and fuzzy logic methods. J Intell Fuzzy Syst 29:2639–2651CrossRefGoogle Scholar
  63. Zhao ZH, Bai ZH, Xiao XL, Mei HJ (2006) The diagenetic and mineralization of the rich alkali igneous rocks in northern China. Xinjiang Geological Press, Beijing 2006:1–302 (in Chinese)Google Scholar
  64. Zuo RG, Carranza EJM (2011) Support vector machine: a tool for mapping mineral prospectivity. Comput Geosci 37:1967–1975CrossRefGoogle Scholar
  65. Zuo R, Wang J (2016) Fractal/multifractal modeling of geochemical data: a review. J Geochem Explor 164:33–41CrossRefGoogle Scholar
  66. Zuo R, Zhang ZJ, Zhang DJ et al (2014) Evaluation of uncertainty in mineral prospectivity mapping due to missing evidence: a case study with skarn-type Fe deposits in southwestern Fujian Province, China. Ore Geol Rev 71:502–515CrossRefGoogle Scholar

Copyright information

© Springer-Verlag GmbH Germany, part of Springer Nature 2018

Authors and Affiliations

  1. 1.State Key Laboratory of Desert and Oasis Ecology, Xinjiang Institute of Ecology and GeographyChinese Academy of SciencesXinjiangChina
  2. 2.Xinjiang Research Center for Mineral Resources, Xinjiang Institute of Ecology and GeographyChinese Academy of SciencesXinjiangChina
  3. 3.Xinjiang Key Laboratory of Mineral Resources and Digital GeologyXinjiangChina
  4. 4.Key Laboratory of Coastal Zone Environmental Processes and Ecological Remediation, Yantai Institute of Coastal Zone Research, Chinese Academy of SciencesYantaiChina
  5. 5.University of Chinese Academy of SciencesBeijingChina

Personalised recommendations