1.No. 1 Geological Team of Shandong Provincial Bureau of Geology and Mineral Resources (Shandong No. 1 Institute of Geology and Mineral Resources Exploration),Jinan,Shandong 250100,China
2.Technology Innovation Center of Geological Information,MNR Shandong Subdivision,Jinan,Shandong 250100,China
3.Shandong Engineering Research Center of Rich Iron Ore Exploration and Development Technology,Jinan,Shandong 250100,China
In recent years, the application of machine learning technology in mineral prospecting has significantly improved the efficiency of mineral resource exploration. However, insufficient model interpretability has become one of the core challenges restricting its large-scale application and implementation. This study systematically reviews the cutting-edge progress of model interpretability research, discusses the research significance of improving interpretability in mineral prospecting and its necessity and urgency in geological science, and proposes methods and recommendations for enhancing the credibility of prediction results derived from different machine learning algorithms at various stages. Existing research primarily revolves around three dimensions: focusing on geological prior knowledge; utilizing visualization techniques to reveal relationships and weight values between global/local outputs and input features; and developing diverse hybrid interpretation frameworks. Nevertheless, critical challenges remain in mineral prospecting applications, including difficulties in decoding high-dimensional nonlinear relationships, inherent complexity of machine learning architectures, and opaque mechanisms of multi-source heterogeneous data fusion. There is an urgent need to develop novel explanatory paradigms that balance computational efficiency with logical consistency. Based on those approaches, this paper proposes specific recommendations such as extending the space-time framework, emphasizing loss functions and visualization techniques, and reasonably select evaluation metrics. The future development direction should prioritize building intelligent prediction systems with enhanced end-to-end interpretability throughout the modeling process, to enhance the logical confidence of prediction frameworks built on different environment, thereby facilitating the optimization and innovation of machine learning models in synergy with frontier metallogenic theories.
KONGY, CHENG, LIUB, et al. 3D mineral prospectivity mapping of Zaozigou gold deposit, west Qinling, China: Machine learning-based mineral prediction[J]. Minerals, 2022, 12(11): 1361.
ZUOR, XIONGY, WANGZ, et al. A new generation of artificial intelligence algorithms for mineral prospectivity mapping[J]. Natural Resources Research, 2023, 32(5): 1859 - 1869.
[11]
RODRIGUEZ-GALIANOV, SANCHEZ-CASTILLOM, CHICA-OLMOM, et al. Machine learning predictive models for mineral prospectivity: An evaluation of neural networks, random forest, regression trees and support vector machines[J]. Ore Geology Reviews, 2015, 71: 804 - 818.
ALGAIARIM, HOSSAINM, PETROVSKIA, et al. Applications of artificial intelligence in geothermal resource exploration: A review[J]. Deep Underground Science and Engineering, 2024, 3(3): 269 - 285.
[16]
SHARMAA, SINGHP K, CHANDRAR. SMOTified-GAN for class imbalanced pattern classification problems[J]. IEEE Access, 2022, 10: 30655 - 30665.
GUD, LIY, JIANGF, et al. VINet: A visually interpretable image diagnosis network[J]. IEEE Transactions on Multimedia, 2020, 22(7): 1720 - 1729.
[21]
YANGC, RANGARAJANA, RANKAS. Global model interpretation via recursive partitioning[C]//2018 IEEE 4th International Conference on Data Science and Systems (HPCC/SmartCity/DSS). IEEE, 2018: 1563 - 1570.
WANGD, QUEKC, NGG S. Bank failure prediction using an accurate and interpretable neural fuzzy inference system[J]. AI Communications, 2016, 29(4): 477 - 495.
[24]
SHAOH, WANGL, CHENR, et al. Safety-enhanced autonomous driving using interpretable sensor fusion transformer[C]//Conference on Robot Learning. PMLR, 2023: 726 - 737.
WANGY, COINTN, MANSURE T, et al. Leveraging domain expertise in machine learning for critical metal prospecting in the Oslo Rift: a case study for Fe-Ti-P-rare earth element mineralization[J]. Minerals, 2024, 14(4): 377.
[31]
ZHANGZ, WANGG, CARRANZAE J M, et al. An uncertainty-quantification machine learning framework for data-driven three-dimensional mineral prospectivity mapping[J]. Natural Resources Research, 2024, 33(4): 1393 - 1411.
SHIRAZIA, HEZARKHANIA, SHIRAZYA, et al. Geochemical modeling of copper mineralization using geostatistical and machine learning algorithms in the Sahlabad area, Iran[J]. Minerals, 2023, 13(9): 1133.
[37]
LIPTONZ C. The mythos of model interpretability[J]. ACM Queue, 2018, 16(3): 31 - 57.
[38]
XIAY, FENGY. Prospects of Artificial Intelligence Application in the High-Quality Development of the Geological Exploration Industry[C]//2023 International Conference on Management Innovation and Economy Development (MIED 2023). Atlantis Press, 2023: 94 - 101.
[39]
ZHANGB, XUZ, WEIX, et al. Deep subsurface pseudo-lithostratigraphic modeling based on three-dimensional convolutional neural network (3D CNN) using inversed geophysical properties and shallow subsurface geological model[J]. Lithosphere, 2024(1): lithosphere_2023_273.
[40]
WASHIZAKIH, YOSHIOKAN. AI Security Continuum: Concept and Challenges[C]//2024 IEEE/ACM 3rd International Conference on AI Engineering-Software Engineering for AI (CAIN). Lisbon, Portugal: IEEE, 2024.
[41]
LUOZ, FARAHBAKHSHE, HORES, et al. An explainable semi-supervised deep learning framework for mineral prospectivity mapping: DEEP-SEAM v1.0[J]. EGUsphere, 2025, 2025: 1 - 43.
[42]
SHAWK O, GOÏTAK, GERMAINM. Prospectivity mapping of heavy mineral ore deposits based upon machine-learning algorithms: Columbite-tantalite deposits in West-Central Côte d'Ivoire[J]. Minerals, 2022, 12(11): 1453.
JIANGS, ZHENGY, SOLOMATINED. Improving AI system awareness of geoscience knowledge: Symbiotic integration of physical approaches and deep learning[J]. Geophysical Research Letters, 2020, 47(13): e2020GL088229.
[47]
MUMUNIA, MUMUNIF. Automated data processing and feature engineering for deep learning and big data applications: a survey[J]. Journal of Information and Intelligence, 2025, 3(2): 113 - 153.
KOSTS, RHEINBACHO, SCHAEBENH. Using logistic regression model selection towards interpretable machine learning in mineral prospectivity modeling[J]. Geochemistry, 2021, 81(4): 125826.
[62]
TANGJ, ZHANGH, BAIR, et al. Mineral prospectivity mapping for exploration targeting of porphyry Cu-polymetallic deposits based on machine learning algorithms, remote sensing and multi-source geo-information[J]. Minerals, 2025, 15(10): 1050.
JIAR, LVY, WANGG, et al. A stacking methodology of machine learning for 3D geological modeling with geological-geophysical datasets, Laochang Sn camp, Gejiu (China)[J]. Computers & Geosciences, 2021, 151: 104754.
[67]
ZHANGQ, ZHOUY, HEJ, et al. A review on global cooperation network in the interdisciplinary research of geochemistry combined with artificial intelligence[J]. Minerals, 2023, 13(10): 1332.
[68]
YANGF, ZUOR, XIONGY, et al. Dual-branch convolutional neural network and its post hoc interpretability for mapping mineral prospectivity[J]. Mathematical Geosciences, 2024, 56(7): 1487 - 1515.
[69]
YINS, LIN, XIAOK, et al. A dual-channel iterative method integrating semi-supervised self-training and interpretable deep learning models for mineral prospectivity prediction[J]. Natural Resources Research, 2025, 34(6): 2953 - 2971.
DONGY L, ZHANGZ J. Deep forest modeling: an interpretable deep learning method for mineral prospectivity mapping[J]. Journal of Geophysical Research: Machine Learning and Computation, 2024, 1(4): e2024JH000311.
[77]
LIZ, MAX, XINH. Feature engineering of machine-learning chemisorption models for catalyst design[J]. Catalysis Today, 2017, 280: 232 - 238.
[78]
FARAHBAKHSHE, MAUGHANJ, MÜLLERR D. Prospectivity modelling of critical mineral deposits using a generative adversarial network with oversampling and positive-unlabelled bagging[J]. Ore Geology Reviews, 2023, 162: 105665.
[79]
ZUOR. Geodata science-based mineral prospectivity mapping: A review[J]. Natural Resources Research, 2020, 29(6): 3415 - 3424.
ZUOR, KREUZERO P, WANGJ, et al. Uncertainties in GIS-based mineral prospectivity mapping: Key types, potential impacts and possible solutions[J]. Natural Resources Research, 2021, 30(5): 3059 - 3079.
[83]
YUX, YUP, WANGK, et al. Data-driven mineral prospectivity mapping based on known deposits using association rules[J]. Natural Resources Research, 2024, 33(3): 1025 - 1048.
[84]
QIUQ, DUANY, MAK, et al. Information extraction and knowledge linkage of geological profiles and related contextual texts from mineral exploration reports for geological knowledge graphs construction[J]. Ore Geology Reviews, 2023, 163: 105739.
[85]
YANQ, XUEL, LIY, et al. Mineral prospectivity mapping integrated with geological map knowledge graph and geochemical data: A case study of gold deposits at Raofeng area, Shaanxi Province[J]. Ore Geology Reviews, 2023, 161: 105651.
CHAKRABORTYS, DATTA GUPTAS, DEVIV, et al. Using rock physics analysis driven feature engineering in ML-based shear slowness prediction using logs of wells from different geological setup[J]. Acta Geophysica, 2024, 72(5): 3237 - 3254.
[95]
ZHUY, ZHONGE, LUZ, et al. Feature engineering for semantic place prediction[J]. Pervasive and Mobile Computing, 2013, 9(6): 772 - 783.
[96]
DAID, XUT, WEIX, et al. Using machine learning and feature engineering to characterize limited material datasets of high-entropy alloys[J]. Computational Materials Science, 2020, 175: 109618.
[97]
WANGQ, MAY, ZHAOK, et al. A comprehensive survey of loss functions in machine learning[J]. Annals of Data Science, 2022, 9(2): 187 - 212.
[98]
KARBALAEIRAMEZANALIA, YOUSEFIF, LENTZD R, et al. Machine learning classification of fertile and barren adakites for refining mineral prospectivity mapping: geochemical insights from the Northern Appalachians, New Brunswick, Canada[J]. Minerals, 2025, 15(4): 372.
[99]
HUJ, SHENL, SUNG. Squeeze-and-excitation networks[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2018: 7132 - 7141.
[100]
SHENGS, WANGY, TIANJ, et al. Knowledge-data collaboration-driven mineral prospectivity prediction with graph attention networks[J]. Minerals, 2025, 15(11): 1164.
ZHOUB L, KHOSLAA, LAPEDRIZAA, et al. Learning deep features for discriminative localization[C]//Proceedings of the 28th IEEE Conference on Computer Vision and Pattern Recognition. Piscataway, NJ: IEEE, 2016: 2921 - 2929.
[112]
FUR, HUQ, DONGX, et al. Axiom-based grad-cam: Towards accurate visualization and explanation of cnns[J]. arXiv preprint arXiv:2020.
[113]
WANGH, WANGZ, DUM, et al. Score-CAM: Score-weighted visual explanations for convolutional neural networks[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops. 2020: 24 - 25.
GHOLAMIH, MOHAMMADIFARA, GOLZARIS, et al. Interpretability of simple RNN and GRU deep learning models used to map land susceptibility to gully erosion[J]. Science of the Total Environment, 2023, 904: 166960.
MCCOYJ T, AURETL. Machine learning applications in minerals processing: A review[J]. Minerals Engineering, 2019, 132: 95 - 109.
[118]
ZHANGS, CARRANZAE J M, FUC, et al. Interpretable machine learning for geochemical anomaly delineation in the Yuanbo Nang district, Gansu Province, China[J]. Minerals, 2024, 14(5): 500.
[119]
MOUN, CARRANZAE J M, WANGG, et al. A framework for data-driven mineral prospectivity mapping with interpretable machine learning and modulated predictive modeling[J]. Natural Resources Research, 2023, 32(6): 2439 - 2462.
ROYT, DAS P, JAGIRDARR, et al. Prediction of mechanical properties of eco-friendly concrete using machine learning algorithms and partial dependence plot analysis[J]. Smart Construction and Sustainable Cities, 2025, 3(1): 2.
[122]
NOHARAY, WAKATAY, NAKASHIMAN. Interpreting Medical Information Using Machine Learning and Individual Conditional Expectation[J]. MedInfo, 2015, 1073.
KONGB, CHENZ. Data-driven EUR modeling and optimization in the liquid-rich Duvernay Formation, western Canada sedimentary basin, Canada[J]. Journal of Petroleum Science and Engineering, 2022, 213: 110352.
[127]
LUNDBERGS M, ERIONG, CHENH, et al. From local explanations to global understanding with explainable AI for trees[J]. Nature Machine Intelligence, 2020, 2(1): 56 - 67.
[128]
LINZ, FENGM, SANTOSC N D, et al. A structured self-attentive sentence embedding[C]//Proceedings of the 5th International Conference on Learning Representations. Toulon, France, 2017: 1 - 15.