| Popis: |
This paper uses a support vector machine (SVM) and back propagation neural network (BPNN) methods to predict the gold in the Dalli deposit in the central province of Iran. For this, the distribution of Au in the ore zone has been predicted after digging some trenches, taking the required samples, and analyzing them. After a building dataset and comprehensive statistical analyses, Au was chosen as an output element modeling, while Cu, Al, Ca, Fe, Ti, and Zn were considered input parameters. Then, the dataset was divided into two groups of training and testing datasets. For this purpose, seventy percent of the datasets were randomly entered into the training process, and the rest of the data were assigned to the test procedure. The correlation coefficients for SVM and BPNN were 94% and 75%, respectively. A comparison of the correlation coefficients revealed that both methods of SVM and BPNN could successfully predict the actual grade of Au. However, SVM was found more reliable and more accurate. |
