Deep Forest Modeling: An Interpretable Deep Learning Method for Mineral Prospectivity Mapping

• Published in: Journal of geophysical research. Machine learning and computation Vol. 1; no. 4
• Main Authors: Dong, Yue‐Lin, Zhang, Zhen‐Jie
• Format: Journal Article
• Language: English
• Published: Wiley 01.12.2024
• Subjects: Cordillera metallogenic belt; Deep Forest; deep learning; mineral prospectivity mapping; porphyry copper
• ISSN: 2993-5210; 2993-5210
• DOI: 10.1029/2024JH000311

Accurate mineral prediction is crucial for reducing costs and uncertainties in mineral discovery and extraction. The use of artificial intelligence and big data has advanced mineral prediction into intelligent forecasting. Machine learning methods have shown significant promise in enhancing outcomes. Currently, neural network‐based approaches dominate deep learning (DL), but they lack interpretability and have high modeling complexity, making them less effective for complex problems and time‐consuming. Deep Forest, an innovative DL paradigm, addresses these issues by dynamically adjusting complexity and providing importance assessments for predictive factors. This study focuses on the North American Cordillera, known for its rich geological data and potential for porphyry copper deposits (PCDs). Predictions are made using Deep Forest with factors like Euclidean distance between faults and magmatic rock, fault line density, gravity anomalies, and stream‐sediment geochemical data. Deep neural networks, random forest, convolutional neural networks, transformer model and graph convolutional networks are also used for comparison. Deep Forest shows high performance and can avoid the black box problem of DL without relying on other tools in DL, providing a new perspective for the development and application of other non‐neural network DL models for mineral prediction. Feature importance analysis shows that geological structure and magmatism significantly influence PCD prediction. Elevated levels of elements like Al, Co, and Cr in stream sediments help identify mineralization‐related alterations. These findings underscore Deep Forest's capability to accurately and efficiently guide mineral exploration, highlighting its potential as a promising approach for mineral prospecting. Plain Language Summary Accurate mineral prediction is important for reducing costs and uncertainties in finding and extracting minerals. Using artificial intelligence and big data has greatly improved this process. Traditional methods like neural networks are complex and hard to understand. Deep Forest is a new method that adjusts its complexity and identifies important factors for predictions. This study focused on the North American Cordillera, an area with rich geological data and potential for valuable mineral deposits. The prediction considered factors like the distance to faults and magmatic rock, fault line density, gravity anomalies, and various elements in stream sediments. Deep Forest outperformed other methods in accuracy, especially in avoiding the black box problems of deep learning and in identifying important geological features and mineral indicators, and provides a new perspective on other non‐neural network‐based mineral prediction models. The results showed that geological structures and certain elements in sediments are crucial for predicting mineral deposits. Deep Forest proved to be an efficient and accurate tool for guiding mineral exploration, making it a promising approach for finding new mineral sources. Key Points Prediction of porphyry copper deposits (PCDs) in the Cordillera region of North America using Deep Forest Deep Forest has the advantages of depth and interpretability and is an available method for mineral prediction Deep Forest offers exploration recommendations for PCDs