Abstract
In order to determine whether it is desirable to quantify mineral-deposit models further, a test of the ability of a probabilistic neural network to classify deposits into types based on mineralogy was conducted. Presence or absence of ore and alteration mineralogy in well-typed deposits were used to train the network. To reduce the number of minerals considered, the analyzed data were restricted to minerals present in at least 20% of at least one deposit type. An advantage of this restriction is that single or rare occurrences of minerals did not dominate the results. Probabilistic neural networks can provide mathematically sound confidence measures based on Bayes theorem and are relatively insensitive to outliers. Founded on Parzen density estimation, they require no assumptions about distributions of random variables used for classification, even handling multimodal distributions. They train quickly and work as well as, or better than, multiple-layer feedforward networks. Tests were performed with a probabilistic neural network employing a Gaussian kernel and separate sigma weights for each class and each variable. The training set was reduced to the presence or absence of 58 reported minerals in eight deposit types. The training set included: 49 Cyprus massive sulfide deposits; 200 kuroko massive sulfide deposits; 59 Comstock epithermal vein gold districts; 17 quartzalunite epithermal gold deposits; 25 Creede epithermal gold deposits; 28 sedimentary-exhalative zinc-lead deposits; 28 Sado epithermal vein gold deposits; and 100 porphyry copper deposits. The most common training problem was the error of classifying about 27% of Cyprus-type deposits in the training set as kuroko. In independent tests with deposits not used in the training set, 88% of 224 kuroko massive sulfide deposits were classed correctly, 92% of 25 porphyry copper deposits, 78% of 9 Comstock epithermal gold-silver districts, and 83% of six quartzalunite epithermal gold deposits were classed correctly. Across all deposit types, 88% of deposits in the validation dataset were correctly classed. Misclassifications were most common if a deposit was characterized by only a few minerals, e.g., pyrite, chalcopyrite,and sphalerite. The success rate jumped to 98% correctly classed deposits when just two rock types were added. Such a high success rate of the probabilistic neural network suggests that not only should this preliminary test be expanded to include other deposit types, but that other deposit features should be added
Article PDF
Similar content being viewed by others
Avoid common mistakes on your manuscript.
References
Barton, P. B., 1986, Mineralogical index,in Cox, D. P., and Singer, D. A., eds., Minerai deposit models: U.S. Geological Survey Bulletin 1693, p. 318–348.
Berger, B. R., 1986, Descriptive model of epithermal quartz-alunite Au,in Cox, D. P., and Singer, D. A., eds., Mineral deposit models: U.S. Geological Survey Bulletin 1693, p. 158.
Briskey, J. A., 1986, Descriptive model of sedimentary exhalative Zn-Pb,in Cox, D. P., and Singer, D. A., eds., Mineral deposit models: U.S. Geological Survey Bulletin 1693, p. 211.
Cox, D. P., 1986, Descriptive model of porphyry Cu,in Cox, D. P., and Singer, D. A., eds., Mineral deposit models: U.S. Geological Survey Bulletin 1693, p. 76.
Cox, D. P., and Singer, D. A., eds., 1986, Mineral deposit models: U.S. Geological Survey Bulletin 1693, 379 p.
Fukunaga, Keinosuke, 1990, Introduction to statistical pattern rec- ognition: second edition, Academic Press, Inc., San Diego, CA, 591 p.
McCammon, R. B., 1992, Numerical mineral deposit models,in Bliss, J. D., ed., Developments in deposit modeling U.S. Geo- logical Survey Bulletin 2004, p. 6–12.
Masters, Timothy, 1995, Advanced algorithms for neural networks: a C++ sourcebook: John Wiley & Sons, New York, 425 p.
Mosier, D. L., Singer, D. A., and Salem, B. B., 1983, Geologic and grade-tonnage information on volcanic-hosted copper- zinc-lead massive sulfide deposits: U.S. Geological Survey Open-File Report 83-89, 78 p.
Mosier, D. L., Berger, B. R., and Singer, D. A., 1986 a, Descriptive model of Sado epithermal veins,in Cox, D. P. and Singer, D. A., eds., Mineral deposit models: U.S. Geological Survey Bulletin 1693, p. 154.
Mosier, D. L., Menzie, W. D., and Kleinhampl, F. J., 1986 b, Geologic and grade-tonnage information on Tertiary epither- mal precious- and base-metal vein districts associated with volcanic rocks: U.S. Geological Survey Bulletin 1666, 39 p.
Mosier, D. L., Singer, D. A., and Berger, B. R., 1986 c Descriptive model of Comstock epithermal veins,in Cox, D. P. and Singer, D. A., eds., Mineral deposit models: U.S. Geological Survey Bulletin 1693, p. 150.
Mosier, D. L., Sato, Takeo, Page, N. J., Singer, D. A., and Berger, B. R., 1986 d, Descriptive model of Creede epithermal veins, in Cox, D. P., and Singer, D. A., eds., Mineral deposit models: U.S. Geological Survey Bulletin 1693, p. 145–146.
Parzen, E., 1962, On estimation of a probability density function and mode: Annals of Mathematical Statistics, v. 33, p. 1065–1076.
Quinlan, J. R., 1986, Induction of decision trees: Machine Learning, v. 1, n. 1, p. 81–106.
Singer, D. A., 1986 a, Descriptive model of Cyprus massive sulfide,in Cox, D. P. and Singer, D. A., eds., Mineral deposit models: U.S. Geological Survey Bulletin 1693, p. 131–133.
inger, D. A., 1986 b, Descriptive model of kuroko massive sulfide,in Cox, D. P. and Singer, D. A., eds., Mineral deposit models: U.S. Geological Survey Bulletin 1693, p. 189–194.
Singer, D. A., and Kouda, Ryoichi, 1996, Application of a feedfor- ward neural network in the search for kuroko deposits in the Hokuroku District, Japan: Mathematical Geology, p. 1017- 1023.
Singer, D. A., 1993, Development of grade and tonnage models for different deposit types,in Kirkham, R. V., Sinclair, R. V., Thorpe, W. D., and Duke, J. M., eds., Mineral deposit model- ing: Geological Association Canada Special Paper 40, p. 21–30.
Specht, Donald, 1990, Probabilistic neural networks: Neural Net- works, v. 3, p. 109–118.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Singer, D.A., Kouda, R. Classification of mineral deposits into types using mineralogy with a probabilistic neural network. Nonrenewable Resources 6, 27–32 (1997). https://doi.org/10.1007/BF02816922
Received:
Accepted:
Issue Date:
DOI: https://doi.org/10.1007/BF02816922