An Artificial Neural Network Method for Mineral Prospectivity Mapping: A Comparison with Fuzzy Logic and Bayesian Probability Methods

  • Warick M. Brown
  • David I. Groves
  • Tamás D. Gedeon
Part of the Modern Approaches in Geophysics book series (MAGE, volume 21)


A multilayer perceptron (MLP) neural network is used to combine multi-source exploration data in a Geographic Information System (GIS) database and produce a mineral prospectivity map for gold deposits in the Tenterfield 1:100,000 sheet area, NSW, Australia. Statistical and probability measures of map quality indicate that the neural network method performs better than the empirical weights-of-evidence (Bayesian probability) and conceptual fuzzy logic methods in the generation of the mineral prospectivity map. The neural network has several important advantages over existing methods, including the ability to respond to critical combinations of parameters, rather than automatically increasing the prospectivity estimate in response to every favourable parameter.


Geographic Infonnation System Fuzzy Logic Gold Deposit Fuzzy Membership Thematic Layer 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. Agterberg, F. P., 1974, Automatic contouring of geological maps to detect target areas for mineral exploration: J. Math. Geol., 6, 373–395.CrossRefGoogle Scholar
  2. An, P., Moon, W. M., and Rencz, A., 1991, Application of fuzzy set theory to integrated mineral exploration: Can. J. Expl. Geophys., 27, 1–11.Google Scholar
  3. Bampton, M. D., 1988, Alteration and mineralisation of the southern part of the Stanthorpe Adamellite, near Tenterfield, New South Wales: B.Sc. thesis, Univ. of Sydney. (unpubl.).Google Scholar
  4. Barnes, R. G., 1993, Use of GIS and image processing for enhancing geological investigations. Case study: Tenterfield 1:100,000 sheet area, northern NSW: Geological Survey Report No. GS1993/290. Geol. Surv. N.S.W.Google Scholar
  5. Barnes, R. G., Henley, H. F., and Henley, J. E., 1995, Exploration data package for the Tenterfield and Coaldale 1:100,000 sheet areas: Geological Survey Report No. GSI995/004. Geol. Surv. N.S.W. Benediktsson, J. A., Swain, P. H., and Ersoy, O. K., 1993, Conjugate-gradient neural networks in classification of multisource and very-high dimensional remote sensing data: Internat. J. Remote Sensing, 14, 2883–2903.Google Scholar
  6. Bonham-Carter, G. F., 1994, Geographic information systems for geoscientists: modelling with GIS, Computer Methods in the Geosciences, 13, Pergamon Press, Inc.Google Scholar
  7. Bonham-Carter, G. F., Agterberg, F. P., and Wright, D. F., 1988, Integration of geological datasets for gold exploration in Nova Scotia: Photo. Eng. and Rem. Sens., 54, 1585–1592.Google Scholar
  8. Bonham-Carter, G. F., Agterberg, F. P., and Wright, D. F., 1989, Weights of evidence: a new approach to mapping mineral potential, in Agterberg, F. P., and Bonham-Carter, G. F., Eds., Statistical Applications in the Earth Sciences: Geol. Surv. Can. Paper 89–9, Geol. Surv. Can., 171–183.Google Scholar
  9. Brown, W. M., Taylor, G. R., Jusmady, Groves, D. I., and Knox-Robinson, C. M., 1998, Application of artificial neural networks to prospectivity analysis in a GIS environment: a comparison with statistical and fuzzy logic methods for Au and Sn deposits of the Tenterfield area, NSW, in 1e Australian Geological Convention, Geological Society of Australia, Abstracts, 49, 57.Google Scholar
  10. Brown, W. M., Gedeon, T. D., and Barnes, R. G., 1999, The use of a multilayer feedforward neural network for mineral prospectivity mapping, in Gedeon, T., Wong, P., Halgamuge, S., Kasabov, N., Nauck, D., and Fukushima, K., Eds., ICONIP’99, Proceedings of the 6“’ International Conference of Neural Information Processing: 1, 160–165, IEEE, Piscataway.Google Scholar
  11. Bruzzone, L., Conese, C., Maselli, F., and Roll, F., 1997, Multisource classification of complex rural areas by statistical and neural-network approaches: Photo. Eng. and Rem. Sens., 63, 523–533.Google Scholar
  12. Clare, A., Bolt, G., and Daneel, G., 1997, The use of neural analysis as an exploration tool, in Proceedings of the 12th International Conference on Applied Geologic Sensing, 2, Environmental Research Institute of Michigan (ERIM), 199–205.Google Scholar
  13. Cucuzza, J., and Goode, A. D. T., 1998, Australian mineral exploration research, in Berkman, D. A. and MacKenzie, D. H. Eds., Geology of Australian and Papua New Guinean mineral deposits: Australasian Institute of Mining and Metallurgy Mono. 22, 53–60.Google Scholar
  14. Decateur, S. E., 1989, Application of neural networks to terrain classification: IEEE INNS International Joint Conference on Neural Networks, 1, 283–288.CrossRefGoogle Scholar
  15. Eddy, B. G., Bonham-Carter, G. F., and Jefferson, C. W., 1995, Mineral resource assessment of the Parry Islands, high Arctic, Canada: a GIS-based fuzzy logic model, in Proceedings Canadian Conference on GIS, CD ROM Session C3, Paper 4.Google Scholar
  16. Harris, D., and Pan, G., 1999, Mineral favourability mapping: a comparison of artificial neural networks, logistic regression, and discriminant analysis: Natural Resources Research, 8, 93–109.Google Scholar
  17. Homik, K., 1991, Approximation capabilities of multilayer feedforward networks: Neural Networks, 4, 251–257.Google Scholar
  18. Jeffrey, W., and Rossner, R., 1986, Optimization algorithms: simulated annealing and neural network processing: Astrophysical Journal, 310, 473–481.Google Scholar
  19. Jusmady, and Taylor, G. R., 1997, Geological Data Integration and Processing System (GEODIPS) software package Version 1. 03: Department of Applied Geology, Univ. of New South Wales.Google Scholar
  20. Kasabov, N.K., 1996, Foundations of neural networks, fuzzy logic, and knowledge engineering: M.I.T. Press. Katz, S. S., 1991, Emulating the Prospector expert system with a raster GIS: Comput. and Geosci., 17, 1033 1050.Google Scholar
  21. Legge, P. J., 1995, Geoscience 1994 and beyond: thoughts on geology and exploration for world-class deposits: Austral. J. Earth Sciences, 42, 1–10.CrossRefGoogle Scholar
  22. Lusted, L. B., 1968, Introduction to medical decision making: Charles Thomas.Google Scholar
  23. Masters, T., 1993, Practical neural network recipes in C++: Academic Press, Inc.Google Scholar
  24. McClelland, J. L., and Rumelhart, D. E., 1988, Explorations in parallel distributed processing: M.I.T. Press.Google Scholar
  25. Moon, W, M. 1990, Integration of geophysical and geological data using evidential belief function: IEEE Trans. Geosci. Remote Sensing, 28, 711–720.Google Scholar
  26. Pan, G., and Harris, D. P., 1992, Estimating a fâvorability equation for the integration of geodata and selection of mineral exploration targets: Mathematical Geology, 24, 177–201.Google Scholar
  27. Reddy, R. K. T., Agterberg, F. P., and Bonham-Carter, G. F., 1991, Application of GIS-based logistic models to base-metal potential mapping in Snow Lake area, in Proceedings of Canadian Conference on GIS, 607618.Google Scholar
  28. Ritter, N. D., and Hepner, G. F., 1990, Application of an artificial neural network to land-cover classification of thematic mapper imagery: Comput. and Geosci., 16, 873–880.Google Scholar
  29. Ritter, N. D., Logan, T. L., and Bryant, N. A., 1988, Integration of neural network technologies with geographic information systems, in Thomas, H. F., Ed., GIS; integrating technology and geoscience applications ( Denver ): National Academy of Science, 102–103.Google Scholar
  30. Rummelhart, D. E., Hinton, G. E., and Williams, R. J., 1986, Learning representations by back-propagating errors: Nature, 323, 533–536.Google Scholar
  31. Sinclair, A. J., and Woodsmith, G. I., 1970, Multiple regression as a method of estimating exploration potential in an area near Terrace, British Columbia: Econ. Geol., 65, 998–1003.CrossRefGoogle Scholar
  32. Singer, D. A., and Kouda, R., 1988, Integrating spatial and frequency information in the search for Kuroko deposits of the Hokuroku District, Japan: Econ. Geol., 83, 18–29.CrossRefGoogle Scholar
  33. Singer, D. A., and Kouda, R., 1996, Application of a feedforward neural network in the search for Kuroko deposits in the Hokuroku District, Japan: Mathematical Geology, 28, 1017–1023.CrossRefGoogle Scholar
  34. Taylor, G. R., 1997, Geoscience GIS: Desktop and Prospectivity Mapping: Short Course at the Key Centre for Mines, July 7–11, 1997, Univ. of New South Wales.Google Scholar
  35. Wang, C., Venkatesh, S. S., and Judd, J. S., 1994, Optimal stopping and effective machine complexity in learning, in Cowan, J. D., and Tesauro, G., and Alspector, J., Eds., Advances in neural information processing systems ( Denver, 1993 ): Morgan Kaufmann Publ., 303–310.Google Scholar
  36. Werbos, P., 1974, Beyond regression; new tools for prediction and analysis in the behavioural sciences: Ph.D. thesis, Harvard Univ.Google Scholar
  37. Woodall, R., 1994, Empiricism and concept in successful mineral exploration: Austral. J. Earth Sciences, 41, 1–10.CrossRefGoogle Scholar
  38. Wybom, L. A. I., Gallagher, R., and Raymond, 0., 1995, Using GIS for mineral potential evaluation in areas with few known mineral occurrences, in Second National Forum on GIS in the Geosciences–Forum Proceedings: Australian Geological Survey Organisation, AGSO Record 1995 /46, 199–211.Google Scholar
  39. Zaknich, A., 1999, Artificial Neural Networks: a research dominated engineering discipline: Centre for Intelligent Information Processing Systems (CIIPS), Department of Electrical and Electronic Engineering, Univ. of Western Australia, (unpublished).Google Scholar
  40. Zimmermann, H. J., 1985, Fuzzy set theory and its applications: Kluwer Academic Publ.Google Scholar

Copyright information

© Springer Science+Business Media Dordrecht 2003

Authors and Affiliations

  • Warick M. Brown
    • 1
  • David I. Groves
    • 1
  • Tamás D. Gedeon
    • 2
  1. 1.Centre for Global Metallogeny, Department of Geology and GeophysicsUniversity of Western AustraliaPerthAustralia
  2. 2.School of Information TechnologyMurdoch UniversityPerthAustralia

Personalised recommendations