Intelligent mineral prospectivity mapping based on the multi-view graph learning

View at publisher Summarise with VIRAT
Authors: Cheng-jie GONG, Bingli Liu, Changjie Cao, Yunhui Kong, Cheng Li, Zhengyao Wang, Zhongli Zhou
Cited by: 0
References: 38

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

针对复杂成矿区多源地学信息耦合关系复杂、已知矿化样本有限及预测结果稳定性不足等问题,以西藏拉萨地区为研究对象,提出一种融合多视图图建模与自监督学习的成矿远景预测方法。该方法将空间邻接关系、地球化学属性相似性及反映构造与岩浆活动背景的连续指标统一构建为多视图图结构,并在残差式GraphSAGE框架中引入跨视图一致性自监督约束,实现对弱标注样本结构信息的充分挖掘。基于少量监督样本引导的半监督学习机制,模型能够稳定刻画区域成矿潜势分布特征。实验结果表明,所提方法在AUPRC和AUROC等指标上优于对比模型,预测结果在空间上与主要控矿构造、岩浆岩分布及多元素地球化学异常具有良好一致性。研究结果表明,该方法在弱标注复杂成矿区的成矿远景评价与靶区优选中具有较高应用潜力。

Fields of Study & Subjects

Geochemistry and Geologic Mapping Geological Modeling and Analysis Machine Learning in Materials Science

Publication details

Journal / source
Bulletin of Mineralogy Petrology and Geochemistry
Volume
45
Issue
3
Pages
760–778
Publication date
23 March 2026
Language
EN