PH
Gold Prospectivity Mapping: Random Forest Vs. Xgboost
Published 8/2026
Created by Dr. Yusuf Abdullahi Musa
MP4 | Video: h264, 1920x1080 | Audio: AAC, 44.1 KHz, 2 Ch
Level: Beginner | Genre: eLearning | Language: English | Duration: 17 Lectures ( 2h 55m ) | Size: 2.8 GB
Master machine learning for mineral exploration using aeromagnetic and radiometric data to delineate gold targets.
What you'll learn
Build end-to-end Python machine learning workflows for gold prospectivity mapping. Process and integrate high-resolution aeromagnetic and airborne radiometric datasets. How to extract, rank, and compare feature importance to identify which geophysical layers contribute most to gold deposit targeting. How to statistically evaluate and compare RF vs. XGB oost performance using cross-validation, confusion matrices and F1-scores. Implement, tune, and compare two top-performing ensemble algorithms: Random Forest and XGB =oost. Convert raw model probabilities into actionable, 4-class gold prospectivity maps (Very Low, Low, Moderate, High). How to export publication-ready 2D prospectivity maps and vector/raster target files directly for GIS integration and drill-targeting decisions.Requirements
Basic understanding of geology, geophysics, or spatial data concepts. Introductory knowledge of Python programming. A computer running Windows, or macOS with Python 3.x installed (Jupyter Notebook or VS Code recommended).Description
Welcome toGold Prospectivity Mapping: Random Forest vs. XGB oost Course-a practical, step-by-step project-based course to modern data-driven mineral exploration!
Traditional mineral exploration can be time-consuming, expensive, and associated with significant uncertainty when selecting areas for follow-up investigation and drilling. Machine learning provides a powerful approach for integrating multiple geoscientific datasets and identifying spatial patterns associated with mineralization. In this course, you will learn how to use machine learning to integrate airborne magnetic and radiometric data with known gold occurrence information to generate predictive gold prospectivity maps.
In this step-by-step course, you will learn how to transform raw geophysical datasets into spatial target maps using Oasis Montaj, ArcMap and Python software and two industry-favorite ensemble algorithms:Random Forest(RF) andExtreme Gradient Boosting (XGB oost).
What Makes This Course Unique?
Hands-on Geophysical Data Integration: Work directly with real-world aeromagnetic grids, radiometric channels (), and known gold occurrence points.Side-by-Side Model Benchmarking: Train, hyperparameter-tune, and directly compare RF and XGB oost models to determine which algorithm performs best on your exploration dataset.Practical Threshold Optimization: Learn how to apply GridSearch and probability thresholding to segment continuous outputs into clear, 4-class prospectivity zones.Industry-Ready Deliverables: Gain the exact workflows needed to produce vector target outputs and GIS-compatible maps used by mining companies and research institutions worldwide.Whether you are looking to upgrade your geoscientific skills, add machine learning to your geoscience toolkit, or complete an academic research project, this course provides the step-by-step guidance you need to succeed.
Who this course is for
Geologists & Exploration Geophysicists looking to integrate machine learning and AI into their spatial targeting workflows. GIS Specialists & Remote Sensing Analysts interested in applied mineral prospectivity modeling. Earth Science Researchers & Students working on data-driven geology, economic geology, or geophysical evaluation papers.Homepage