北京测绘2025，Vol.39Issue(11)：1700-1706,7.DOI:10.19580/j.cnki.1007-3000.2025.11.022
融合LLMs和MCP技术的数据分析与知识服务智能体研究与应用
Research and application of data analysis and knowledge service intelligent agent integrating LLMs and MCP technology
摘要
Abstract
To address issues in traditional spatiotemporal data analysis,such as the complexity of professional software operations,difficulties in converting unstructured requirements,and inefficient repetitive tasks,this study explored new approaches to drive the intelligent transformation of industries using the powerful logical reasoning,multimodal output,and instruction execution capabilities of large language models(LLMs).This study integrated LLMs with model context protocol(MCP)technology to develop a data analysis and knowledge service agent(DAKS-Agent).This intelligent agent aims to automate the mapping of user natural language instructions to professional geographic information system(GIS)software tools.The test results are as follows.① MCP,as a standardized intermediate protocol,achieves seamless integration between LLMs and GIS professional software.② It supports the standardized processing of spatiotemporal data,spatial analysis,thematic mapping,and efficient output of decision support results.The modular design of DAKS-Agent supports dynamic expansion of third-party data sources and algorithm libraries,effectively addressing challenges such as multi-source heterogeneous data governance,data value transformation,and precise knowledge service delivery.③ The DAKS-Agent framework proposed in this study provides a practical and feasible new approach for the intelligent and universal application of GIS analysis,significantly improving analysis efficiency and usability.关键词
大语言模型(LLMs)/模型上下文协议(MCP)/智能体/数据分析/知识服务Key words
large language model(LLM)/model context protocol(MCP)/intelligent agent/data analysis/knowledge service分类
天文与地球科学引用本文复制引用
陈星宇,曾泳谕,徐佳沅,王思敏..融合LLMs和MCP技术的数据分析与知识服务智能体研究与应用[J].北京测绘,2025,39(11):1700-1706,7.基金项目
自然资源部华南热带亚热带地区自然资源监测重点实验室开放基金(2022NRM006) （2022NRM006）
