The increasing costs and uncertainties associated with the early stages of mineral exploration highlight the need for advanced analytical approaches to improve the efficiency of exploration processes. This study investigates the application of machine learning algorithms to enhance the efficiency of mineral exploration through the separation of geochemical anomalies. The Maraki area in Hormozgan Province, Iran, was selected as the case study, and regional geochemical data were analyzed using a combination of statistical methods and machine learning models. descriptive statistical analysis and correlation analysis were conducted to examine the characteristics of the dataset and identify relationships among geochemical elements. Subsequently, classical statistical approaches, including the standard deviation method and the median absolute deviation, were applied to distinguish background values from anomalous concentrations. In the next stage, the Random Forest algorithm was employed as an effective machine learning method to identify complex patterns and nonlinear relationships among geochemical elements. The results indicate that the Random Forest model performs better than traditional statistical methods in identifying areas with high mineralization potential. The probability maps generated by the model also show a strong spatial correspondence with the geological and structural trends of the study area. the findings suggest that integrating machine learning techniques with conventional geological and geochemical analyses can significantly improve the accuracy of exploration data interpretation, reduce uncertainty in the early stages of exploration, and enhance the economic efficiency of mineral exploration activities. These results highlight the potential of data driven approaches for improving decision making and planning in mineral exploration.