GeoGraphRAG: A graph-based retrieval-augmented generation approach for empowering large language models in automated geospatial modeling

Geospatial modeling aims to integrate multiple geospatial data sources and geoprocessing functions based on user application demands to address complex geospatial challenges. While large language models (LLMs) have shown remarkable capabilities in semantic understanding and task planning, their limi...

Full description

Saved in:
Bibliographic Details
Main Authors: Jianyuan Liang, Shuyang Hou, Haoyue Jiao, Yaxian Qing, Anqi Zhao, Zhangxiao Shen, Longgang Xiang, Huayi Wu
Format: Article
Language:English
Published: Elsevier 2025-08-01
Series:International Journal of Applied Earth Observations and Geoinformation
Subjects:
Online Access:http://www.sciencedirect.com/science/article/pii/S1569843225003590
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850036146567905280
author Jianyuan Liang
Shuyang Hou
Haoyue Jiao
Yaxian Qing
Anqi Zhao
Zhangxiao Shen
Longgang Xiang
Huayi Wu
author_facet Jianyuan Liang
Shuyang Hou
Haoyue Jiao
Yaxian Qing
Anqi Zhao
Zhangxiao Shen
Longgang Xiang
Huayi Wu
author_sort Jianyuan Liang
collection DOAJ
description Geospatial modeling aims to integrate multiple geospatial data sources and geoprocessing functions based on user application demands to address complex geospatial challenges. While large language models (LLMs) have shown remarkable capabilities in semantic understanding and task planning, their limited geospatial domain knowledge still hinders their efficiency in geospatial modeling. Therefore, there is an urgent need to integrate geospatial knowledge with the inherent semantic understanding capabilities of LLMs, enabling their effective application in specialized geospatial domains. In this study, we introduce GeoGraphRAG, a graph-based retrieval-augmented generation (RAG) approach tailored for geospatial modeling. Specifically, GeoGraphRAG utilizes LLMs as intelligent agents to identify user demands, retrieve relevant subgraphs from external graph knowledge base, and injects both structural and semantic information into the LLM. Through the graph-driven solution planning phase, GeoGraphRAG automatically generates modeling solutions and geospatial code in response to specific user demands. The proposed method is evaluated against different baseline approaches across various LLMs based on the benchmark dataset designed for geospatial modeling, and a representative application scenario is presented to illustrate its practical effectiveness. Our experimental results demonstrate that the injection of graph-structured knowledge improves both the efficiency and interpretability of LLM-based geospatial modeling, thereby confirming the effectiveness of our proposed approach. GeoGraphRAG represents the first attempt to introduce graph-based RAG into geospatial modeling and provides novel insights for future research and practical applications in this field.
format Article
id doaj-art-4725698106d74be49946f231ea7ec45c
institution DOAJ
issn 1569-8432
language English
publishDate 2025-08-01
publisher Elsevier
record_format Article
series International Journal of Applied Earth Observations and Geoinformation
spelling doaj-art-4725698106d74be49946f231ea7ec45c2025-08-20T02:57:17ZengElsevierInternational Journal of Applied Earth Observations and Geoinformation1569-84322025-08-0114210471210.1016/j.jag.2025.104712GeoGraphRAG: A graph-based retrieval-augmented generation approach for empowering large language models in automated geospatial modelingJianyuan Liang0Shuyang Hou1Haoyue Jiao2Yaxian Qing3Anqi Zhao4Zhangxiao Shen5Longgang Xiang6Huayi Wu7State Key Laboratory of Information Engineering in Surveying, Mapping and Remote Sensing, Wuhan University, Wuhan, ChinaState Key Laboratory of Information Engineering in Surveying, Mapping and Remote Sensing, Wuhan University, Wuhan, ChinaState Key Laboratory of Information Engineering in Surveying, Mapping and Remote Sensing, Wuhan University, Wuhan, ChinaState Key Laboratory of Information Engineering in Surveying, Mapping and Remote Sensing, Wuhan University, Wuhan, ChinaState Key Laboratory of Information Engineering in Surveying, Mapping and Remote Sensing, Wuhan University, Wuhan, ChinaState Key Laboratory of Information Engineering in Surveying, Mapping and Remote Sensing, Wuhan University, Wuhan, ChinaState Key Laboratory of Information Engineering in Surveying, Mapping and Remote Sensing, Wuhan University, Wuhan, ChinaCorresponding author.; State Key Laboratory of Information Engineering in Surveying, Mapping and Remote Sensing, Wuhan University, Wuhan, ChinaGeospatial modeling aims to integrate multiple geospatial data sources and geoprocessing functions based on user application demands to address complex geospatial challenges. While large language models (LLMs) have shown remarkable capabilities in semantic understanding and task planning, their limited geospatial domain knowledge still hinders their efficiency in geospatial modeling. Therefore, there is an urgent need to integrate geospatial knowledge with the inherent semantic understanding capabilities of LLMs, enabling their effective application in specialized geospatial domains. In this study, we introduce GeoGraphRAG, a graph-based retrieval-augmented generation (RAG) approach tailored for geospatial modeling. Specifically, GeoGraphRAG utilizes LLMs as intelligent agents to identify user demands, retrieve relevant subgraphs from external graph knowledge base, and injects both structural and semantic information into the LLM. Through the graph-driven solution planning phase, GeoGraphRAG automatically generates modeling solutions and geospatial code in response to specific user demands. The proposed method is evaluated against different baseline approaches across various LLMs based on the benchmark dataset designed for geospatial modeling, and a representative application scenario is presented to illustrate its practical effectiveness. Our experimental results demonstrate that the injection of graph-structured knowledge improves both the efficiency and interpretability of LLM-based geospatial modeling, thereby confirming the effectiveness of our proposed approach. GeoGraphRAG represents the first attempt to introduce graph-based RAG into geospatial modeling and provides novel insights for future research and practical applications in this field.http://www.sciencedirect.com/science/article/pii/S1569843225003590Large Language Model (LLM)Geospatial modelingDomain knowledge injectionGraph-based retrieval-augmented generationIntelligent generation
spellingShingle Jianyuan Liang
Shuyang Hou
Haoyue Jiao
Yaxian Qing
Anqi Zhao
Zhangxiao Shen
Longgang Xiang
Huayi Wu
GeoGraphRAG: A graph-based retrieval-augmented generation approach for empowering large language models in automated geospatial modeling
International Journal of Applied Earth Observations and Geoinformation
Large Language Model (LLM)
Geospatial modeling
Domain knowledge injection
Graph-based retrieval-augmented generation
Intelligent generation
title GeoGraphRAG: A graph-based retrieval-augmented generation approach for empowering large language models in automated geospatial modeling
title_full GeoGraphRAG: A graph-based retrieval-augmented generation approach for empowering large language models in automated geospatial modeling
title_fullStr GeoGraphRAG: A graph-based retrieval-augmented generation approach for empowering large language models in automated geospatial modeling
title_full_unstemmed GeoGraphRAG: A graph-based retrieval-augmented generation approach for empowering large language models in automated geospatial modeling
title_short GeoGraphRAG: A graph-based retrieval-augmented generation approach for empowering large language models in automated geospatial modeling
title_sort geographrag a graph based retrieval augmented generation approach for empowering large language models in automated geospatial modeling
topic Large Language Model (LLM)
Geospatial modeling
Domain knowledge injection
Graph-based retrieval-augmented generation
Intelligent generation
url http://www.sciencedirect.com/science/article/pii/S1569843225003590
work_keys_str_mv AT jianyuanliang geographragagraphbasedretrievalaugmentedgenerationapproachforempoweringlargelanguagemodelsinautomatedgeospatialmodeling
AT shuyanghou geographragagraphbasedretrievalaugmentedgenerationapproachforempoweringlargelanguagemodelsinautomatedgeospatialmodeling
AT haoyuejiao geographragagraphbasedretrievalaugmentedgenerationapproachforempoweringlargelanguagemodelsinautomatedgeospatialmodeling
AT yaxianqing geographragagraphbasedretrievalaugmentedgenerationapproachforempoweringlargelanguagemodelsinautomatedgeospatialmodeling
AT anqizhao geographragagraphbasedretrievalaugmentedgenerationapproachforempoweringlargelanguagemodelsinautomatedgeospatialmodeling
AT zhangxiaoshen geographragagraphbasedretrievalaugmentedgenerationapproachforempoweringlargelanguagemodelsinautomatedgeospatialmodeling
AT longgangxiang geographragagraphbasedretrievalaugmentedgenerationapproachforempoweringlargelanguagemodelsinautomatedgeospatialmodeling
AT huayiwu geographragagraphbasedretrievalaugmentedgenerationapproachforempoweringlargelanguagemodelsinautomatedgeospatialmodeling