知识图谱与图神经网络助力智慧城市的可持续发展  

Knowledge Graphs and Graph Neural Networks Drive Sustainable Development in Smart Cities

作　　者：常普 赵轶鸣 云曼玉 刘永明 赵转哲[1,3] CHANG Pu;ZHAO Yiming;YUN Manyu;LIU Yongming;ZHAO Zhuanzhe(School of Artificial Intelligence,Anhui Polytechnic University,Wuhu,Anhui,241000,China;Wuhu Ceprei Information Industry Technology Research Institute Co.,Ltd.,Wuhu,Anhui,241000,China;Anhui Provincial Key Laboratory of Discipline Co-construction on Intelligent Equipment Quality and Reliability,Wuhu 241000,China)

机构地区：[1]安徽工程大学人工智能学院,安徽芜湖241000 [2]芜湖赛宝信息产业技术研究院有限公司,安徽芜湖241000 [3]智能装备质量与可靠性安徽省学科共建重点实验室,安徽芜湖241000

出　　处：《智能城市应用》2025年第2期124-128,共5页Smart City Application

基　　金：安徽省重点实验室开放课课题(APELDE2023A005,AIMTEEL202201);安徽省质量工程项目(2023sdxx043,2023xjzlts040);安徽工程大学校级质量工程项目(2022jyxm02);安徽工程大学研究生“机器学习”线上示范课程。

摘　　要：本研究旨在通过知识图谱与图神经网络(GNN)的结合,构建一种创新的预测模型,以提升智慧城市中可持续发展目标(SDGs)的预测与决策能力。研究首先介绍了智慧城市与SDGs的背景,强调了传统方法在应对数据孤岛和复杂系统交互等挑战时的局限性。接着详细阐述了知识图谱与GNN的技术原理,包括数据整合、关系捕捉和动态适应能力。通过模拟案例W市,研究展示了模型在资源分配优化中的应用效果。结果表明,该模型能够有效提升资源分配的效率和SDGs的实现水平,为政策制定者提供了科学依据。研究的结论是,知识图谱与GNN的结合为智慧城市的可持续发展提供了新的技术手段,未来在技术进步和应用场景拓展的推动下,将在城市发展中发挥更加重要的作用。This study aims to construct an innovative predictive model by integrating knowledge graphs with Graph Neural Networks(GNNs)to enhance the prediction and decision-making capabilities for Sustainable Development Goals(SDGs)in smart cities.The research begins by introducing the background of smart cities and SDGs,highlighting the limitations of traditional methods in addressing challenges such as data silos and complex system interactions.It then elaborates on the technical principles of knowledge graphs and GNNs,including data integration,relationship capture,and dynamic adaptability.Through a simulated case study of City W,the research demonstrates the model's effectiveness in optimizing resource allocation.The results indicate that the model significantly improves the efficiency of resource allocation and the achievement of SDGs,providing a scientific basis for policymakers.The study concludes that the integration of knowledge graphs and GNNs offers a new technological approach for the sustainable development of smart cities,and with advancements in technology and the expansion of application scenarios,it will play an increasingly important role in urban development.

关 键 词：智慧城市 可持续发展目标 知识图谱 图神经网络 资源分配优化 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]