GeoChemAD: Benchmarking Unsupervised Geochemical Anomaly Detection for Mineral Exploration

2026-03-13

AI Summary

Existing geochemical anomaly detection methods suffer from limited generalizability and poor reproducibility due to their reliance on single-region and proprietary datasets. To address this, this work introduces GeoChemAD—the first open-source benchmark dataset encompassing multiple regions, sampling sources, and target elements—and proposes GeoChemFormer, a target-element-aware self-supervised Transformer framework. By leveraging self-supervised pretraining to learn robust geochemical representations, GeoChemFormer achieves state-of-the-art and consistently stable performance across all eight subsets of GeoChemAD, significantly outperforming existing unsupervised approaches. This study thus provides a reproducible and highly generalizable solution for geochemical anomaly detection.

Technology Category

• Data Mining & Knowledge Management: Anomaly/Outlier Detection
• Machine Learning: Unsupervised & Self-Supervised Learning
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning

Application Category

• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets

Abstract

Geochemical anomaly detection plays a critical role in mineral exploration as deviations from regional geochemical baselines may indicate mineralization. Existing studies suffer from two key limitations: (1) single region scenarios which limit model generalizability; (2) proprietary datasets, which makes result reproduction unattainable. In this work, we introduce \textbf{GeoChemAD}, an open-source benchmark dataset compiled from government-led geological surveys, covering multiple regions, sampling sources, and target elements. The dataset comprises eight subsets representing diverse spatial scales and sampling conditions. To establish strong baselines, we reproduce and benchmark a range of unsupervised anomaly detection methods, including statistical models, generative and transformer-based approaches. Furthermore, we propose \textbf{GeoChemFormer}, a transformer-based framework that leverages self-supervised pretraining to learn target-element-aware geochemical representations for spatial samples. Extensive experiments demonstrate that GeoChemFormer consistently achieves superior and robust performance across all eight subsets, outperforming existing unsupervised methods in both anomaly detection accuracy and generalization capability. The proposed dataset and framework provide a foundation for reproducible research and future development in this direction.

Problem

• geochemical anomaly detection
• mineral exploration
• model generalizability
• reproducibility
• unsupervised learning

Innovation

• GeoChemAD
• unsupervised anomaly detection
• transformer-based framework
• self-supervised pretraining
• geochemical representation