Geochemical Behavior Investigation Based on K-Means Clustering

One of the most important viewpoints of data mining concerning the analysis of a large amount of data and samples with different characteristics includes clustering which is associated with methods and techniques that have frequently been applied in different fields of researches. A well-known algorithm of clustering is K-means by which the data are divided into K classes based upon a distance criterion. In this study, the K-means method applies for data classification derived from exploration boreholes and surface sampling in the Parkam deposit. The results derived based on the case of borehole data showed that the increasing Cu grade values resulted in a significant increase in Mo grades, a significant decrease in Pb grades followed by an increase, and the Zn grades varying comparable to Pb. With regards to the relationships between these elements it could be concluded that the meteoric waters promoted the mobilization of Pb and Zn from the potassic to the phyllic zone but the meteoric waters were not effective enough to cause the mobilization of Cu, and this element together with Mo remained immobile. Finally, equation of Cu grade estimation is also presented for surface data.