QueryPlot: Generating Geological Evidence Layers using Natural Language Queries for Mineral Exploration
arXiv:2602.17784v1 Announce Type: cross Abstract: Mineral prospectivity mapping requires synthesizing heterogeneous geological knowledge, including textual deposit models and geospatial datasets, to identify regions likely to host specific mineral deposit types. This process is traditionally manual and knowledge-intensive. We present QueryPlot, a semantic retrieval and mapping framework that integrates large-scale geological text corpora with geologic map data using modern Natural Language Processing techniques. We curate descriptive deposit models for over 120 deposit types and transform the State Geologic Map Compilation (SGMC) polygons into structured textual representations. Given a user-defined natural language query, the system encodes both queries and region descriptions using a pretrained embedding model and computes semantic similarity scores to rank and spatially visualize regions as continuous evidence layers. QueryPlot supports compositional querying over deposit character
arXiv:2602.17784v1 Announce Type: cross Abstract: Mineral prospectivity mapping requires synthesizing heterogeneous geological knowledge, including textual deposit models and geospatial datasets, to identify regions likely to host specific mineral deposit types. This process is traditionally manual and knowledge-intensive. We present QueryPlot, a semantic retrieval and mapping framework that integrates large-scale geological text corpora with geologic map data using modern Natural Language Processing techniques. We curate descriptive deposit models for over 120 deposit types and transform the State Geologic Map Compilation (SGMC) polygons into structured textual representations. Given a user-defined natural language query, the system encodes both queries and region descriptions using a pretrained embedding model and computes semantic similarity scores to rank and spatially visualize regions as continuous evidence layers. QueryPlot supports compositional querying over deposit characteristics, enabling aggregation of multiple similarity-derived layers for multi-criteria prospectivity analysis. In a case study on tungsten skarn deposits, we demonstrate that embedding-based retrieval achieves high recall of known occurrences and produces prospective regions that closely align with expert-defined permissive tracts. Furthermore, similarity scores can be incorporated as additional features in supervised learning pipelines, yielding measurable improvements in classification performance. QueryPlot is implemented as a web-based system supporting interactive querying, visualization, and export of GIS-compatible prospectivity layers.To support future research, we have made the source code and datasets used in this study publicly available.
Executive Summary
The article introduces QueryPlot, a semantic retrieval and mapping framework designed to streamline mineral prospectivity mapping by integrating large-scale geological text corpora with geospatial datasets. Utilizing modern Natural Language Processing (NLP) techniques, QueryPlot enables users to generate geological evidence layers through natural language queries. The framework curates descriptive deposit models for over 120 deposit types and transforms geologic map data into structured textual representations. By encoding queries and region descriptions using pretrained embedding models, QueryPlot computes semantic similarity scores to rank and visualize prospective regions. A case study on tungsten skarn deposits demonstrates the framework's effectiveness in achieving high recall of known occurrences and aligning with expert-defined permissive tracts. The system also supports interactive querying, visualization, and export of GIS-compatible prospectivity layers, with the source code and datasets made publicly available for future research.
Key Points
- ▸ QueryPlot integrates geological text corpora with geospatial datasets using NLP techniques.
- ▸ The framework supports compositional querying and multi-criteria prospectivity analysis.
- ▸ A case study on tungsten skarn deposits demonstrates high recall and alignment with expert-defined tracts.
- ▸ QueryPlot is implemented as a web-based system with interactive querying and visualization capabilities.
- ▸ The source code and datasets are publicly available to support future research.
Merits
Innovative Integration of NLP and Geological Data
QueryPlot effectively bridges the gap between textual geological knowledge and spatial data, leveraging advanced NLP techniques to enhance mineral prospectivity mapping.
User-Friendly Interface
The web-based system allows for interactive querying and visualization, making it accessible to a broader range of users, including those with limited technical expertise.
High Recall and Alignment with Expert Knowledge
The case study demonstrates that QueryPlot achieves high recall of known occurrences and aligns closely with expert-defined permissive tracts, validating its effectiveness.
Demerits
Dependence on Quality of Input Data
The effectiveness of QueryPlot is highly dependent on the quality and comprehensiveness of the input geological text corpora and geospatial datasets.
Potential Bias in Embedding Models
The use of pretrained embedding models may introduce biases that could affect the accuracy and reliability of the semantic similarity scores.
Limited Scope of Case Study
The case study focuses on tungsten skarn deposits, and the framework's performance may vary when applied to other types of mineral deposits.
Expert Commentary
QueryPlot represents a significant advancement in the integration of NLP techniques with geological data for mineral prospectivity mapping. The framework's ability to transform complex geological knowledge into actionable insights through natural language queries is a notable achievement. The case study on tungsten skarn deposits provides a robust validation of the framework's effectiveness, demonstrating high recall and alignment with expert knowledge. However, the dependence on the quality of input data and potential biases in embedding models are important considerations that warrant further investigation. The public availability of the source code and datasets is commendable and will undoubtedly support future research in this area. Overall, QueryPlot has the potential to revolutionize mineral exploration by making prospectivity mapping more accessible, efficient, and accurate.
Recommendations
- ✓ Further validation of QueryPlot's performance across a broader range of mineral deposit types is recommended to ensure its generalizability.
- ✓ Investigation into the potential biases in the embedding models and the development of mitigation strategies should be prioritized to enhance the framework's reliability.
