One of the first decisions to be made when building a mineral resource model is the definition of geological/geostatistical domains. Cluster analysis is a set of techniques in machine learning that can be especially suited for this matter. In order to compare different approaches, two clustering algorithms were investigated in this study: k-means and the dual-space clustering algorithm. Choosing the most appropriate method and the number of clusters can be challenging and some metrics are needed to support these decisions, including the validation of the spatial distribution of the clusters, which is not always appropriately discussed in the literature. We introduce the use of correlograms of the indicators for that matter. Although clustering techniques can be robust for an application in resource modelling, expert knowledge is still necessary when applying cluster analysis to resource modeling, since final decisions should not be based solely on statistical indexes, but also on experience. In this paper, the proposed methodology was tested in a three-dimensional dataset related to a phosphate/titanium deposit.
AmbroiseC, DangM, GovaertG.1997. Clustering of spatial data by the EM algorithm.Proceedings of GEOEV I – Geostatistivs for Environmental Applications. p. 493–504. doi: 10.1007/978-94-017-1675-8_40
2.
ArthurD, VassilvitskiiS.2007. K-means++: the advantages of careful seeding.Proceedings of the Annual ACM-SIAM Symposium on Discrete Algorithms. p. 1027–1035; New Orleans, Louisiana. doi:10.5555/1283383.1283494.
CalinskiT, HarabaszJ.1974. A dendrite method for cluster analysis.Commun Stat – Theor M. 3(1):1–27. doi:10.1080/03610927408827101.
5.
DempsterAP, LairdNM, RubinDB.1977. Maximum likelihood from incomplete data via the EM algorithm.J R Stat Soc B (Methodol). 39(1):1–22. doi:10.1111/j.2517-6161.1977.tb01600.x.
6.
D‘UrsoP, VitaleV.2020. A robust hierarchical clustering for georeferenced data.Spat Stat. 35. doi:100407.
7.
FouedjioF.2016a. A hierarchical clustering method for multivariate geostatistical data.Spat Stat. 18:333–351. doi:10.1016/j.spasta.2016.07.003.
8.
FouedjioF.2016b. Discovering spatially contiguous clusters in multivariate geostatistical data through spectral clustering.Proceedings of the Thirteenth International Conference in Machine Learning and Data Mining in Pattern Recognition. Vol. 10086 LNAI, p. 547–557. doi: 10.1007/978-3-319-49586-6_38
9.
FouedjioF.2017a. A spectral clustering approach for multivariate geostatistical data.Int J Data Sci Anal. 4(4):301–312. doi:10.1007/s41060-017-0069-7.
10.
FouedjioF.2017b. A spectral clustering method for large-scale geostatistical datasets.Proceedings of the Thirteenth International Conference in Machine Learning and Data Mining in Pattern Recognition. Vol. 10358 LNAI, p. 248–261. doi: 10.1007/978-3-319-62416-7_18
FouedjioF., HillEJ, LaukampC.2018. Geostatistical clustering as an aid for ore body domaining: case study at the Rocklea Dome channel iron ore deposit, Western Australia.Appl Earth Sci: Transe Inst Min Metall. 127(1):15–29. doi:10.1080/03717453.2017.1415114.
13.
GetisA, OrdJK.1992. The analysis of spatial association by use of distance statistics.Geogr Anal. 24(3):189–206. doi:10.1111/j.1538-4632.1992.tb00261.x.
MacQueenJ.1967. Some methods for classification and analysis of multivariate observations.Proceedings of the fifth Berkeley Symposium on Mathematical Statistics and Probability.Vol. 1, p. 281–296.
16.
MarizC, WagnerL, PilgerG, WhitehouseP, BessinC, LaurentD.2016. Geostatistical clustering, potential modeling and traditional geostatistics: a new coherent workflow.Proceedings of the Sixth International Conference on Innovation in Mine Operations; Santiago, Chile.
17.
MartinR.2019. Data driven decisions of stationarity for improved numerical modeling in geological environments. Edmonton, Canada: University of AlbertaSage UK: London, England.
18.
MartinR, BoisvertJ.2018. Towards justifying unsupervised stationary decisions for geostatistical modeling: ensemble spatial and multivariate clustering with geomodeling specific clustering metrics.Comput Geosci. 120(August):82–96. doi:10.1016/j.cageo.2018.08.005.
19.
ModenaRCC, MoreiraGC, MarquesDM, CostaJFCL.2019. Avaliação de técnicas de agrupamento para definição de domínios estacionários com o auxílio de geoestatística.Proceedings of the twentieth mining symposium at the fifht ABM week; São Paulo. p. 91–100.
20.
MoreiraGC.2020. Análise de agrupamento aplicada à definição de domínios de estimativa para a modelagem de recursos minerais. Porto Alegre, Brazil: Universidade Federal do Rio Grande do SulSage UK: London, England. https://lume.ufrgs.br/handle/10183/212457.
21.
OliverMA, WebsterR.1989. A geostatistical basis for spatial weighting in multivariate classification.Math Geol. 21(1):15–35. doi:10.1007/BF00897238.
22.
OrdJK, GetisA.1995. Local spatial autocorrelation statistics: distributional issues and an application.Geogr Anal. 27:286–306. doi:10.1111/j.1538-4632.1995.tb00912.x.
PradesC.2017. Geostatistics and clustering for geochemical data analysis.Edmonton, Canada: University of AlbertaSage UK: London, England.
25.
RomaryT, OrsF, RivoirardJ, DeraismeJ.2015. Unsupervised classification of multivariate geostatistical data: two algorithms.Comput Geosci. 85(December):96–103. doi:10.1016/j.cageo.2015.05.019.
26.
RomaryT, RivoirardJ, DeraismeJ, QuinonesC, FreulonX.2012. Domaining by clustering multivariate geostatistical data.Geostatistics Oslo. p. 455–466. doi: 10.1007/978-94-007-4153-9_37
27.
RossiME, DeutschCV.2014. Mineral resource estimation. Springer Science & Business MediaSage UK: London, England. doi:10.1007/978-1-4020-5717-5.
28.
RousseeuwP.J.1987. Silhouettes: a graphical aid to the interpretation and validation of cluster analysis.J Comput Appl Math. 20(C):53–65. doi:10.1016/0377-0427(87)90125-7.
29.
ScruccaL.2005. Clustering multivariate spatial data based on local measures of spatial autocorrelation.Quaderni del Dipartimento di Economia, Finanza e Statistica. 20(1):1–25. http://www.ec.unipg.it/DEFS/uploads/spatcluster.pdf.
30.
SoaresA.2000. Geoestatística para as ciências da terra e do ambiente. Lisbon: Instituto Superior TécnicoSage UK: London, England.
31.
SokalRR, SneathPHA.1965. Principles of numerical taxonomy.J Mammal. 46(1):111. doi:10.2307/1377831.
32.
SrivastavaRM, ParkerHM.1989. Robust Measures of Spatial Continuity. Geostatistics. Vol. 4. Dordrecht: SpringerSage UK: London, England; p. 295–308. doi:10.1007/978-94-015-6844-9_22.
33.
StrehlA, GhoshJ.2002. Cluster ensembles – a knowledge reuse framework for combining multiple partitions.J Mach Learn Res. 3:583–617. doi:10.1162/153244303321897735.
