Application of the k-Prototype Clustering Approach for the Definition of Geostatistical Estimation Domains

The definition of geostatistical domains is a stage in the estimation mineral resources, which sample resulting from mining exploration process divided into zones that show homogeneity or minimal variation main element interest grade, having geological and spatial meaning. Its importance lies fact quality techniques, therefore, correct quantification resource, will improve geostatistically stationary areas. present study seeks to define for using non-traditional approach based on k-prototype clustering algorithm. This algorithm k-means paradigm unsupervised machine learning, but it exempt one-time restriction numeric data. latter especially convenient, as allows incorporation categorical variables such attributes grouping. case corresponds hydrothermal gold deposit high sulfidation, located southern zone Peru, where are defined historical record data recovered 131 diamond drill holes 37 trenches. characteristics directly involved were grade (Au), silver (Ag), type alteration, mineralization. results obtained showed with k-prototypes an efficient can be used alternative complement traditional methodology.