Exploration Approaches for Identifying Key Mineral Resource Prospects

This study provides an in-depth review of mineral exploration methodologies, tracing the evolution from traditional approaches to the most recent advancements in machine learning algorithms. Over the past three to four decades, GIS-based methods have been widely used for mineral mapping, demonstrating substantial efficacy. However, the emergence of more advanced methodologies, particularly machine learning (ML) algorithms, has significantly enhanced computer-based mapping in mineral exploration. Notably, Random Forest-a prominent shallow ML algorithm-and Convolution Neural Networks (CNNs)-a key Deep Learning approach-have emerged as powerful tools in this domain. This work highlights the application of ML-based algorithms for determining Lead and Zinc mineral compositions in a study region from which data is collected as in-situ analysis. All three algorithms are analyzed for effectiveness and accuracy.