Tag: mineral prospectivity mapping

New research shows that coarse, smoothed labels in continental-scale AI mineral prospectivity maps artificially inflate model performance while erasing high-probability ...

A validated weighted-overlay prospectivity model, built from resistivity and magnetic data along Ethiopia's western Afar rift margin, captured all known ...

A new framework combining geological constraints and point pattern analysis shows that the choice of non-deposit training locations can dramatically ...

Researchers at the University of Ghana used random forest and decision tree classifiers trained on airborne geophysical data to map ...

Researchers at Shandong University of Technology show that convolutional neural networks, especially AlexNet, can map gold prospectivity at the Jinchang ...

A new dual-masked graph autoencoder called DM-GAE maps mineral prospectivity in Tibet with high accuracy despite scarce labeled deposits, outperforming ...

A self-supervised Vision Transformer framework that fuses multi-scale geological maps with aeromagnetic data has substantially outperformed conventional methods in mapping ...

A new hybrid 3D CNN-graph attention framework with entropy-guided fusion achieved an AUROC of 0.964 and delineated five new exploration ...

A multi-level stacking ensemble of five machine learning models mapped porphyry copper-gold prospectivity in northwest Iran with an AUC of ...

A knowledge–data dual-driven machine learning framework combining deep forest algorithms with orogenic gold mineral system knowledge has improved mineral prospectivity ...