We are excited to invite you to the next seminar of the 2025 Geology and Geophysics Seminar Series, featuring Elnaz Heidari, who is a PhD candidate in School of Geosciences, University of Sydney. Elnaz will be presenting on “Machine Learning Framework for Prospectivity Mapping of Porphyry Mineralisation in the Lachlan Fold Belt, Eastern Australia“. In this engaging talk, she will present a machine learning–driven prospectivity framework using PU Bagging and a Random Forest classifier, integrating geological, geophysical, and remote sensing datasets to predict porphyry mineralisation, sharing innovative methods in geoscience.
Date: November 26, 2025
Time: 11:00 a.m. – 12:00 p.m., Sydney Time
Location: Room 449 (Conference Room), Madsen Building (F09), School of Geosciences
or Online (Join via zoom)
We look forward to seeing you there in person or joining us online!
https://uni-sydney.zoom.us/j/88080833538?from=addon
Machine Learning Framework for Prospectivity Mapping of Porphyry Mineralisation in the Lachlan Fold Belt, Eastern Australia
Abstract
The transition to renewable energy technologies has increased the demand for critical and strategic minerals, such as copper (Cu), molybdenum (Mo). This growing demand highlights the need for more efficient approaches to mineral exploration. Since most near-surface deposits have already been explored or mined, it is crucial to investigate deeper ore deposits. On the other hand, traditional exploration techniques can be time-consuming, costly and sometimes lack accuracy, and mineral exploration is complex and non-linear, requiring technological advancements and human creativity for sustained success. Advanced technologies, such as artificial intelligence (AI) and machine learning (ML), have revolutionized mineral exploration by enabling the efficient use of diverse datasets, increasing accuracy and saving time and money.
Our study uses machine learning–based mineral prospectivity mapping to predict porphyry mineralisation in the Lachlan Fold Belt, Australia. Using a Positive–Unlabelled bagging approach with a Random Forest classifier, the framework handles challenges such as limited negative samples and complex datasets. Integrating geological, geophysical, and remote sensing data with deposit size weighting. This model identifies high-potential zones in this region and ranks them into four exploration priorities. It captures 95% of known mineral occurrences within just 5% of the study area, demonstrating high accuracy. Additionally, our model identifies the most important data layers and features for the modelling process. These features played a significant role in discriminating between mineralised and non-mineralised areas. The generated prospectivity map reveals a clear spatial correlation between areas of high probability and known mineral occurrences while also highlighting several potential greenfield zones for further exploration
Graphical Abstract
