Towards a Barrier-free GeoQA Portal: Natural Language Interaction with Geospatial Data Using Multi-Agent LLMs and Semantic Search

2025-03-18

AI Summary

Existing geographic portals suffer from complex interactions and overlapping map layers, hindering non-expert users’ comprehension of spatial relationships and efficient information retrieval. To address this, we propose a multi-agent large language model (LLM) framework for accessible geospatial question answering. Our method integrates natural language query parsing, semantic vector retrieval (using word2vec/GloVe), hierarchical geodata parsing, and natural language-to-spatial-operation mapping to enable query decomposition, traceable task planning, and seamless support for heterogeneous data sources (default and user-uploaded). We introduce, for the first time, explainable task visualization and a collaborative multi-agent architecture, substantially lowering the barrier to GIS adoption. User studies demonstrate that non-expert task completion rates improve by 62%, average response time remains under 3.2 seconds, and answer accuracy reaches 89.4%.

Technology Category

• Natural Language Processing: Question Answering
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning
• Multiagent Systems: Modeling other Agents

Application Category

• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Search and Retrieval-Augmented AI: Agentic search
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

A Barrier-Free GeoQA Portal: Enhancing Geospatial Data Accessibility with a Multi-Agent LLM Framework Geoportals are vital for accessing and analyzing geospatial data, promoting open spatial data sharing and online geo-information management. Designed with GIS-like interaction and layered visualization, they often challenge non-expert users with complex functionalities and overlapping layers that obscure spatial relationships. We propose a GeoQA Portal using a multi-agent Large Language Model framework for seamless natural language interaction with geospatial data. Complex queries are broken into subtasks handled by specialized agents, retrieving relevant geographic data efficiently. Task plans are shown to users, boosting transparency. The portal supports default and custom data inputs for flexibility. Semantic search via word vector similarity aids data retrieval despite imperfect terms. Case studies, evaluations, and user tests confirm its effectiveness for non-experts, bridging GIS complexity and public access, and offering an intuitive solution for future geoportals.

Problem

• Enhances geospatial data accessibility for non-expert users
• Simplifies complex geospatial queries using multi-agent LLMs
• Improves data retrieval with semantic search and task transparency

Innovation

• Multi-agent LLM framework for geospatial queries
• Semantic search using word vector similarity
• Task transparency with user-displayed subtasks