Department of Mining Engineering, Isfahan University of Technology
Abstract
Dashkasan exploration area contains Sari Gunay and Agh Dagh deposits. The Sari Gunay epithermal gold deposit, with 120 Mt reserve with a mean grade of 2 g/t, is the most important Iranian gold deposit in the world-class. The locus of the gold mineralization has been modelled by geochemical soil data through two classification methods including the discrimination analysis (LDA and QDA) and support vector machine (c-SVM and nu-SVM). The productivity index, sum of the multiplication of block grade by its height, is defined for each surface cell by 25×25 meters. This dependent variable has been classified to background, medium and high productivity indices. Geochemical data overlapping with these cells are used to calculate the classifying functions. 70% of data are used as train and the rest as the test data. The results show that the nu-SVM with 57.9%, LDA with 80% and nu-SVM, c-SVM and LDA with 57.9% accuracy can be applied to divide background, medium and high productivity index areas in the test data, respectively. Therefore, combined discrimination analysis and support vector machine methods can model gold mineralization in Sari Gunay hill. Modelling using all geochemical data show that the gold mineralization in Agh Dagh hill is similar to Sari Gunay hill but less extensive and the connection between the two mineralization is in depth. In the two areas in Sari Gunay hill, one area in Agh Dagh hill and one area in the valley between these two hills are recommended for further drilling.
Tabatabaei, S. H. (2015). Application of discrimination analysis and support vector machine methods for modelling in the epithermal gold deposits in Dashkasan area. Journal of Mining Engineering, 10(28), 53-65.
MLA
Seyed Hassan Tabatabaei. "Application of discrimination analysis and support vector machine methods for modelling in the epithermal gold deposits in Dashkasan area". Journal of Mining Engineering, 10, 28, 2015, 53-65.
HARVARD
Tabatabaei, S. H. (2015). 'Application of discrimination analysis and support vector machine methods for modelling in the epithermal gold deposits in Dashkasan area', Journal of Mining Engineering, 10(28), pp. 53-65.
VANCOUVER
Tabatabaei, S. H. Application of discrimination analysis and support vector machine methods for modelling in the epithermal gold deposits in Dashkasan area. Journal of Mining Engineering, 2015; 10(28): 53-65.
)
