Apriori Algorithm-Based Three-Dimensional Mineral Prospectivity Mapping—An Example from Meiling South Area, Xinjiang, China

Mineral Prospectivity Mapping (MPM) is shifting toward intelligent deep mineralization searches in the era of big data and increasing difficulties surface deposit detection. Comparative analysis two forms prediction based on Apriori algorithm was performed Meiling South mining area eastern Hami region Xinjiang, China. In comparison 1, we use to mine ore-forming information determine voxel positions spatial distance angle analysis. Then, compare determined by with predicted mathematical model conceptual mineralization, these models include Gaussian Naive Bayesian (GNB) Support Vector Machine (SVM). 2, optimal SVM, which trained using elements mineralization. sets new are extracted from original Chi-square methods then input into SVM for training. After obtain results, them results. The preceding produced following (1) Using algorithm, distribution characteristics high low-grade ore bodies association rules between ore-bearing were determined. (2) results GNB displayed corresponding trends voxels identified Apriori, matched mined Apriori. (3) screened chosen have best effect when tested drill holes. Based number accurate (prediction probability greater than 0.5) 6, 5, 1 holes V1, V2, V3, respectively. collective demonstrated that explicit, intuitive, interpretable has a certain reference value refining determination discovering mechanisms laws.