Session Details
Machine learning (ML) and AI are rapidly changing mineral exploration and mining, enabling more objective, data rich decisions. In exploration, ML based prospectivity mapping integrates geology, geophysics, geochemistry and remote sensing to highlight targets that traditional approaches may miss. In operations, computer vision applied to core photos can standardize and accelerate core logging, while robust data cleaning is essential for reliable geochemical interpretation. Recent advances in large language models also create new opportunities and risks for non programmers to build and use ML tools. This course is timely because many mining organizations have large, underused datasets and a skills gap in practical ML/AI methods. It is targeted to technical professionals in mining (geoscientists, mining engineers, and data analysts) but has broad appeal to managers and decision makers seeking to understand realistic AI capabilities and pitfalls.
This 2 day practical course explores how machine learning (ML) and artificial intelligence (AI) can be applied across the mineral exploration and mining value chain. Focusing on real-world examples rather than theory, the course will demonstrate how ML/AI can help integrate diverse geoscientific datasets, improve interpretation and decision-making, and automate or augment technical workflows such as core logging and geochemical analysis. Participants will be introduced to key concepts, common workflows, and typical challenges, as well as emerging tools (including modern AI assistants) and their limitations. Lectures are complemented by guided exercises, giving attendees a foundation to critically evaluate and begin applying ML/AI methods in their own organizations.
Top takeaways:
- Explain key ML/AI concepts and how they apply to mineral exploration and mining workflows
- Outline a practical end to end workflow for using geoscientific data in ML/AI applications (from data preparation to model evaluation)
- Identify common data quality issues in geological and geochemical datasets and describe strategies to mitigate them
- Demonstrate strategies to extract the most from geoscientific datasets (i.e., geochemical, geophysical, structural, remote sensing) and how to approach text-based datasets (e.g. lithology)
- Critically assess the use of modern AI tools (e.g., LLM-based assistants) in mining-related tasks, including typical risks, biases, and governance considerations
Laptop required for course exercises.