The massive accumulation of available multi-modal mineral exploration data for most metallogenic belts worldwide provides abundant information for the discovery of mineral resources. However, managing and analyzing these ever-growing and multidisciplinary mineral exploration data has become increasingly time-consuming and labor-intensive. Artificial intelligence (AI) has demonstrated powerful prediction and knowledge integration capabilities, enabling geologists to efficiently leverage mineral exploration data. This paper reviews publications on state-of-the-art AI applications for ten mineral exploration tasks ranging from data mining to grade and tonnage estimation. These studies are based on expert systems, fuzzy logic, and various machine learning algorithms designed to optimize and improve the workflow of mineral exploration. We recognize that most AI for mineral exploration is data-driven research for now. However, AI models that couple geological knowledge and mineral exploration data will be increasingly favored in this field in the future. This paper also discusses the challenges of AI in mineral exploration research and the implications of future developments associated with novel technologies and practical deployments. Although AI has not yet been extensively tested for practical deployment in mineral exploration, its study execution exhibits the potential to trigger a fundamental research paradigm shift.

中文翻译：

---

人工智能在矿产勘探中的应用：数据科学对未来发展方向的回顾和展望


全球大多数成矿带的可用多模式矿产勘探数据的大量积累为矿产资源的发现提供了丰富的信息。然而，管理和分析这些不断增长的多学科矿产勘探数据变得越来越耗时和劳动密集。人工智能 （AI） 展示了强大的预测和知识集成能力，使地质学家能够有效地利用矿产勘探数据。本文回顾了有关十项矿产勘探任务的最新 AI 应用的出版物，这些任务包括数据挖掘、品位和吨位估算等。这些研究基于专家系统、模糊逻辑和各种机器学习算法，旨在优化和改进矿产勘探的工作流程。我们认识到，目前大多数用于矿产勘探的 AI 都是数据驱动的研究。然而，未来该领域将越来越受到该领域青睐的 AI 模型。本文还讨论了人工智能在矿产勘探研究中面临的挑战，以及与新技术和实际部署相关的未来发展的影响。尽管人工智能尚未在矿产勘探中进行实际部署的广泛测试，但其研究执行显示出引发基础研究范式转变的潜力。