统一知识库检索

全局检索与知识浏览

跨论文、博客、数据集线索、项目和工具统一检索。检索结果可以继续跳转到独立问答页,做语义追问和来源核验。

检索条件
用当前检索词继续问答
当前展示 24 条,共 1081
筛选后的知识库结果
首页/全局检索
论文
arXiv
SpatialIntelligence
ComplexNetwork
异质空间复杂网络中意见动态的稳定边界
Stable Boundaries of Opinion Dynamics in Heterogeneous Spatial Complex Networks

我们研究了在几何非均匀随机图(GIRGs)上的多数投票意见动态,该模型是空间复杂网络的强大表征。与经典粗化动态中通常达成全局共识的情况不同,我们的模拟表明,足够大且局部集中的意见区域不会消失。相反,它们趋于稳定,导致竞争性意见的持续共存。为理解这种被抑制的粗化机制,我们构建并分析了一个两个意见区域界面的可处理均值场模型。我们的主要理论结果在均值场分析中严格证明了界面轮廓存在一个稳定且非平凡的极限分布。这表明意见边界处于静止状态,为复杂网络几何结构如何在社会系统中支持稳健的意见多样性提供了数学解释。

Mats Bierwirth, Johannes Lengler
2026/03/10
论文
Sustainable Cities and Society
PublisherJournal
UrbanComputing
基于复杂网络视角评估与提升城市地铁系统的内涝韧性
Assessing and enhancing the waterlogging resilience of urban subway systems from a complex-network perspective

出版日期:2026年10月1日 来源:《可持续城市与社会》(Sustainable Cities and Society),第149卷 作者:徐飞、宋长青、王浩、高彤、方德林

Sustainable Cities and Society
论文
Cities
PublisherJournal
Platform
从表征到预见?三维城市模拟与深度学习如何重塑理论地理学与计量地理学
From representation to foresight? How 3D urban simulations and deep learning are reshaping theoretical and quantitative geography

出版日期:2026年12月;来源:《Cities》,第179卷;作者:Igor Agbossou、Jean-Philippe Antoni

论文
Landscape and Urban Planning
UrbanCompLab
GeoSimulation
CoCA:城市群“土地-人口-经济”系统空间协同仿真与未来预测
CoCA: Spatial cooperative simulation and future prediction of “land-population-economy” in urban agglomerations
Zeng, Chenglong, Yao, Yao, Liu, Jiayao
2025/01/01
论文
Sustainable Cities and Society
PublisherJournal
UrbanComputing
如何构建与阻断热暴露网络以缓解城市热岛效应?一种结合形态学空间格局分析(MSPA)与复杂网络理论的新方法
How to construct and block a heat exposure network to mitigate urban heat island effect? A novel approach using MSPA and complex network theory

出版日期:2026年5月15日;来源:《可持续城市与社会》(Sustainable Cities and Society),第142卷;作者:熊瑶、许攀岳、朱志鹏、李寅

Sustainable Cities and Society
论文
Sustainable Cities and Society
PublisherJournal
基于居民参与的可持续社区更新:一项基于复杂网络的演化动力学研究
Sustainable community renewal with resident participation: A study on evolutionary dynamics based on complex networks

出版日期:2026年5月15日;来源:《可持续城市与社会》(Sustainable Cities and Society),第142卷;作者:刘飞、徐国亮、李明

Sustainable Cities and Society
论文
Computers, Environment and Urban Systems
UrbanCompLab
UrbanTraffic
融合蚁群优化与复杂网络分析评估“社区开放”政策对缓解中国大城市交通拥堵的影响
Estimating the effects of “community opening” policy on alleviating traffic congestion in large Chinese cities by integrating ant colony optimization and complex network analyses

《计算机、环境与城市系统》;第70卷;第163-174页;2018年发表;出版商:Elsevier

Yao, Yao, Hong, Ye, Wu, Daiqiang
2018/01/01
论文
arXiv
GeoAI
GIS
GenWorld:面向可扩展大语言模型智能体研究的经验驱动型城市仿真基础设施
GenWorld: Empirically Grounded Urban Simulation Infrastructure for Scalable LLM-Agent Studies

大语言模型(LLM)智能体仿真面临联合的接地性(grounding)与可扩展性问题:智能体需在反映真实城市约束的环境中行动,但对城市规模人口直接调用在线LLM在计算上不可行。本文提出GenWorld——一种经验驱动的城市仿真基础设施,整合了建筑级合成城市、结构化的智能体-环境交互接口,以及将LLM生成的决策信号离线编译为查找式策略(lookup policies)以实现可扩展仿真的机制。在以日本东广岛市为参考的实例化中,GenWorld基于人口普查与地理空间数据构建了196,608名合成居民,通过人口普查汇总数据验证其人口统计一致性,并利用YJMob100K手机移动数据作为通勤距离诊断依据。我们通过三个可复现案例展示该基础设施:全城工作日仿真、工作日与周末行为对比、以及含可审计重规划轨迹的预警响应扰动实验。这些案例表明GenWorld可作为支持接地性与可扩展性兼具的LLM智能体研究的可复现平台;而针对交通、疏散或政策效果的校准化预测仍属未来工作。

Gen Li, Jieyuan Lan, Pengcheng Xu
2026/06/26
论文
arXiv
Trajectory
Mobility
当合理不等于真实:评估基于大语言模型的urban simulation中人类移动行为的真实性
When Plausible Is Not Realistic: Evaluating Human Mobility in LLM-Based Urban Simulation

基于大语言模型(LLM)的生成式智能体正日益被用于城市模拟器,但其是否能再现经验上真实的人类移动模式,抑或仅生成表面合理的移动叙事,仍不明确。我们提出一种验证框架,用于将LLM驱动的城市模拟器中生成式智能体的移动行为与真实世界移动数据进行比对评估。该框架涵盖移动定律、时间节律、网络模体、语义活动转移以及行为移动画像等维度。我们利用大巴黎地区与上海的数据集,在多个移动真实性维度上评估AgentSociety与CitySim。分析表明,叙事合理性与经验移动真实性之间存在显著差距。尽管这些模拟器能在一定程度上捕捉高层语义活动分布,却难以再现核心时空约束,包括真实的行程长度分布、起讫点(OD)流、停留时长及转移动态。我们进一步发现,真实的移动多样性在默认提示配置下不稳定,可能需依赖显式的画像感知初始化。为支持可复现的评估,我们还贡献了一套可扩展、开源的LLM驱动基础设施,支持区域尺度地图生成、可观测性增强的模拟、移动度量计算及交通模拟。本研究强调了对LLM驱动城市模拟器开展严格经验验证的必要性,并提供了构建更真实、更可复现城市模拟系统的实用工具。

Gustavo H. Santos, Aline Carneiro Viana, Thiago H. Silva
2026/06/12
论文
arXiv
ComplexNetwork
UrbanTraffic
一种新颖的引力-拟拉普拉斯方法用于识别复杂网络中的关键节点
A Novel Gravity-Quasi-Laplacian Approach to Identifying Influential Nodes in Complex Networks

识别复杂网络中的关键节点是一项基础性挑战,在社交网络分析、通信基础设施、交通系统和信息网络等领域具有广泛应用。现有排序方法通常依赖于度数、k-壳指数及邻域连通性等结构特征的组合来估计节点的重要性。然而,许多此类方法存在若干关键局限性,包括准确性不足、对影响力相近节点的区分能力(分辨率)较低、依赖可调参数,以及计算复杂度高,从而限制了其在大规模或真实网络中的实用性。本研究提出一种新的排序框架,将拟拉普拉斯结构度量与受引力启发的聚合过程相结合。其核心思想是仅利用度数和k-壳指数这两种简单但信息丰富的属性,构建每个节点结构角色的增强表征,并通过短程相互作用机制评估其局部影响力。所提方法无需可调参数、具备可解释性且计算高效,仅需一个固定的小引力半径(R=3),因而适用于大规模且异构的网络。在九个真实网络上开展的实验表明,相较于八种前沿方法,该框架在准确性、分辨率和计算简洁性方面均持续优于现有技术。结果凸显了引力-拟拉普拉斯范式作为识别复杂网络关键节点的一种可靠且可扩展工具的有效性。

Shima Esfandiari, Seyed Mostafa Fakhrahmad
2026/07/26
论文
Landscape and Urban Planning
UrbanCompLab
GeoSimulation
HashGAT-VCA:一种结合哈希函数与图注意力网络的向量细胞自动机模型用于城市土地利用变化模拟
HashGAT-VCA: A vector cellular automata model with hash function and graph attention network for urban land-use change simulation
Guan, Qingfeng, Li, Jianfeng, Zhai, Yaqian
2024/01/01
论文
International Journal of Geographical Information Science
PublisherJournal
UrbanComputing
基于增强人类移动性的元胞自动机(HME-CA)模型对城市群开展经济与人口模拟
Economic and demographic simulations in urban agglomerations using a human mobility-enhanced cellular automata (HME-CA) model
Wei Tu Zhuoyuan Bao Xiaojuan Liu Yatao Zhang Wei Gao Mingxiao Li a Department of Urban Informatics, School of Architecture and Urban Planning, Shenzhen University, Shenzhen, Guangdong Province, P.R. Chinab School of Architecture and Urban Planning, Guangdong Key Laboratory for Urban Informatics, Shenzhen University, Shenzhen, P.R. Chinac Ministry of Nature Resources, Key Laboratory for Geo-Environmental Monitoring of Great Bay Area, Shenzhen, P.R. Chinad State Key Laboratory of Subtropical Building and Urban Science, Shenzhen, P.R. Chinae Department of Geography, University College London, London, United Kingdomf College of Civil and Transportation Engineering, Shenzhen University, Shenzhen, Guangdong Province, P.R. ChinaWei Tu is currently a Professor in the Department of Urban Spatial Information Engineering, Shenzhen University. His research interests include automatics recognition of human activity mobility from multi-source urban data, trajectory modeling, and analysis and optimization. His contributions to this paper include conceptualization, methodology, investigation, supervision, original draft writing, manuscript editing, and funding acquisition.Zhuoyuan Bao is a Master’s student at the Department of Urban Informatics, Shenzhen University. His research interests include urban simulation and spatiotemporal modeling. His contributions to this paper include software, validation, formal analysis, data curation, visualization.Xiaojuan Liu is a Postdoctoral Research Fellow in geographical sciences at Shenzhen University. Her research interests focus on spatiotemporal modelling across land use change, urban development, and ecological effects. Her contributions to this paper include data curation, visualization, manuscript review and editing.Yatao Zhang is a Senior Research Fellow at the Department of Geography, University College London. His research sits at the intersection of GeoAI and urban analytics, with a specific focus on spatiotemporal context modeling across urban transportation, land use, and geodemographics. His contributions to this paper include conceptualization, data curation, manuscript review and editing.Wei Gao is a Master’s student at the Department of Urban Informatics, Shenzhen University. His research lies at the inter-section of urban simulation and spatial modelling. His contributions to this paper include software, data curation, validation, and formal analysis.Mingxiao Li is an Assistant Professor at the College of Civil and Transportation Engineering, Shenzhen University. His research interests focus on urban computing, human mobility, GeoAI, and spatiotemporal data mining. His contributions to this paper include software, data curation, visualization.
2026/05/13
论文
Computers, Environment and Urban Systems
UrbanCompLab
GIS
基于斑块生成土地利用模拟(PLUS)模型理解可持续土地扩展的驱动因素:以中国武汉为例
Understanding the drivers of sustainable land expansion using a patch-generating land use simulation (PLUS) model: A case study in Wuhan, China

《计算机、环境与城市系统》;第85卷;第101569页;2021年发表;出版机构Pergamon

Liang, Xun, Guan, Qingfeng, Clarke, Keith C
2021/01/01
论文
International Journal of Geographical Information Science
UrbanCompLab
GIS
粤港澳大湾区土地利用-人口-经济空间协同模拟
Spatial cooperative simulation of land use-population-economy in the Greater Bay Area, China

快速城市化给可持续发展目标带来了巨大挑战,如资源过度开发和人口激增。传统的元胞自动机(CA)模型被广泛用于独立模拟空间特征的变化,例如土地利用、人口分布、经济产出等。然而,大多数CA模型依赖历史数据作为静态驱动因子来预测未来情景,忽略了发展过程中多种特征之间的相互耦合影响。为解决这一问题,本研究提出一种空间协同模拟(SCS)方法,用于模拟土地利用、人口与经济的变化。SCS方法首先通过独立的CA模型获取各特征的初始状态;随后,将其他两个特征的模拟结果作为动态更新的驱动因子,而非静态的历史数据,以捕捉多特征在发展过程中的相互耦合效应。该过程迭代进行,直至各特征变化收敛,最终输出模拟结果。在粤港澳大湾区的模拟实验表明,SCS方法能够有效捕捉多要素的协同演化过程,优于基准方法。该方法具备预测未来发展趋势的能力,有助于支持空间规划与基础设施协同发展。

Tu, Wei, Gao, Wei, Li, Mingxiao
2024/01/01
论文
arXiv
ComplexNetwork
一种空间网络解释:城市幂律层级结构
A spatial network explanation for a hierarchy of urban power laws

该模型为空间经济学中若干经验观测现象提供了理论解释。通过将系统表征为由固定大小土地单元构成的复杂网络,各单元间通过港口活动间的贸易相互连接,从而导出城市规模与土地价值分布等高阶模式。模型预测了经验观测到的土地价值密度空间分布以及城市集群价格的空间分布。为将土地价值与人口(一种常见可观测量)关联起来,我们还证明:在城市集群层面,累积土地价值与人口之间存在线性关系。

Claes Andersson, Alexander Hellervik, Kristian Lindgren
2003/06/17
论文
arXiv preprint arXiv:1705.05651
UrbanCompLab
GeoSimulation
融合元胞自动机与随机森林的中国城市扩张与耕地损失模拟
Simulation of urban expansion and farmland loss in China by integrating cellular automata and random forest

基于地理信息系统(GIS)和元胞自动机技术,构建了城乡扩展现象的空间模拟模型。该模型揭示了建成区向周边农村地区扩张以满足现有城区发展需求的过程。通过结合地理信息系统方法,利用土地适宜性分析法,依据地形坡度、距市中心距离等地理与可达性因素,确定了优化发展区域的土地利用变化概率。

Yao, Yao, Liu, Xiaoping, Zhang, Dachuan
2017/01/01
论文
Cities
UrbanCompLab
GeoSimulation
基于需求驱动的随机森林-元胞自动机模型融合集成引力场模型模拟城市扩张
Simulating urban expansion by incorporating an integrated gravitational field model into a demand-driven random forest-cellular automata model
Lv, Jianjun, Wang, Yifan, Liang, Xun
2021/01/01
论文
International Journal of Digital Earth
PublisherJournal
Platform
UniGCA:一种支持栅格与矢量数据的通用图结构元胞自动机框架,用于城市增长模拟
UniGCA: a universal graph cellular automata framework for both raster- and vector-based urban growth simulation

城市增长模拟对可持续空间规划与政策制定至关重要。元胞自动机(CA)是其中应用最广泛的方法之一;现有模型可分为基于栅格(RCA)和基于矢量(VCA)两类范式,二者在结构上的差异制约了模型的整合性与可迁移性。本文提出一种通用图结构元胞自动机(UniGCA)框架,以同等支持上述两种范式,并应对空间结构设计与空间交互建模中的挑战。基于UniGCA框架,进一步构建了分区图U-Net元胞自动机(PGUN-CA)模型作为其实用实现。PGUN-CA采用优化的图存储策略与多级图划分方法,以克服大规模城市模拟中的内存与计算约束;同时融合图U-Net与注意力机制,以刻画复杂的空间交互关系。以上海市(栅格数据)与澳大利亚松河流域(矢量数据)为案例的研究表明,PGUN-CA较基线模型精度更高,其指标(Figure of Merit, FoM)分别提升6.96%与9.41%。这些结果凸显了UniGCA框架的通用性与可扩展性,可为异构数据类型下的大规模城市增长模拟提供实用、灵活的解决方案,并为城市规划与可持续发展提供坚实支撑。

Xiaoyu Chen Xuefeng Guan Qingyang Xu Lijun Jiang Huayi Wu a State Key Laboratory of Information Engineering in Surveying, Mapping and Remote Sensing, Wuhan University, Wuhan, Chinab Department of Civil and Environmental Engineering, University of California, Berkeley, CA, USA
2026/05/15
论文
Cities
UrbanCompLab
GIS
Temporal-VCA:基于耦合时间数据与向量细胞自动机的城市土地利用变化模拟
Temporal-VCA: Simulating urban land use change using coupled temporal data and vector cellular automata
Yao, Yao, Zhou, Kun, Liu, Chenxi
2024/01/01
论文
arXiv
Trajectory
Mobility
CityReal:基于大规模大语言模型智能体的人本对齐城市行为与城市动态仿真框架
CityReal: Human-Aligned Urban Behavior and City Dynamics Simulation with Large-Scale LLM Agents

大规模城市仿真在社会科学、交通安全及交通政策研究中具有关键作用。近期研究表明,将大语言模型(LLM)作为智能体进行提示(prompting),可在城市尺度上生成类人的日常行为序列。然而,现有方法通常依赖少样本提示(few-shot prompting),导致智能体主要复现LLM自身的行为先验,而非目标人群的真实行为模式。本文提出CityReal——一种面向人本对齐的城市仿真模块化框架。CityReal将智能体建模为意图驱动的决策者,使其追求连贯的出行与活动计划,而非孤立的逐步选择;并依据经验与约束持续学习习惯与偏好,实现动态适应。为提升群体层面的真实性,我们为各行为模块学习文本适配器(textual adapters),使智能体决策与观测到的人群统计特征对齐。实验表明,CityReal在微观个体行为与宏观群体模式两个层面均显著提升了对真实人类行为的拟合度。该框架可扩展至数万个智能体规模,支持在不同城市情景下分析人群密度、场所热度、出行流以及居民福祉等指标,为城市仿真与预测提供可扩展的试验平台。

Nicolas Bougie, Xiaotong Ye, Narimasa Watanabe
2026/07/08
论文
Sustainable Cities and Society
PublisherJournal
UrbanComputing
利用机器学习与多情景模拟增强湖环型城市群中冷热岛网络的热韧性
Enhancing thermal resilience of heat–cold island networks in lake-ring urban agglomerations using machine learning and multi-scenario simulations

出版日期:2026年5月14日在线发表 来源:《可持续城市与社会》(Sustainable Cities and Society) 作者:熊素文、杨帆、范航源、蒋亚东、舒凯、朱宁静

Sustainable Cities and Society
论文
Landscape and urban planning
UrbanCompLab
GeoSimulation
Delineating multi-scenario urban growth boundaries with a CA-based FLUS model and morphological method

Landscape and urban planning;卷 177;页码 47-63;发表于 2018 年;出版机构 Elsevier。

Liang, Xun, Liu, Xiaoping, Li, Xia
2018/01/01
论文
International Journal of Geographical Information Science
PublisherJournal
Platform
一种用于分离数字连通性与地理溢出效应的多网络空间误差工作流:基于性传播感染的证据
A multi-network spatial error workflow for disentangling digital connectivity and geographic spillovers: evidence from sexually transmitted infections
Fengrui Jing Tong Li Zhenlong Li M. Naser Lessani Jinjing Hu Guanhao He Jianxiong Hu Tao Liu Wenjun Ma a Department of Public Health and Preventive Medicine, School of Medicine, Jinan University, Guangzhou, Chinab Geoinformation and Big Data Research Lab, Department of Geography, The Pennsylvania State University, University Park, PA, USAc School of Geography and Planning, Sun Yat-sen University, Guangzhou, ChinaFengrui Jing is a Research & Teaching-Track Lecturer/Assistant Professor at the Department of Public Health and Preventive Medicine at Jinan University. His research interests include geospatial data and human health, and theoretical and methodological development pertaining to GIS & spatial statistics. His contributions to this paper include conceptualization, methodology, code development, investigation, supervision, original draft writing, manuscript editing, and funding acquisition. Email: [email protected] Li is a Master’s student at the Department of Public Health and Preventive Medicine at Jinan University. Her research interests include big data analysis & statistics. Her contributions to this paper include software, validation, formal analysis, and data curation.Zhenlong Li is an Associate Professor in the Department of Geography at The Pennsylvania State University. His primary research field is GIScience with a focus on geospatial big data analytics, spatial computing, and geospatial AI with applications to human mobility and public health. He contributed to conceptualization of the research idea, manuscript review, and data curation.M. Naser Lessani is a PhD candidate in the Department of Geography at The Pennsylvania State University. His primary research interests are geospatial data analytics and GIS. His contributions to this paper include conceptualization, code review, manuscript review and editing.Jinjing Hu is a Postdoc Researcher in the School of Geography and Planning at Sun Yat-sen University. Her research lies at the inter-section of urban simulation and climate modelling. Her contributions to this paper include software, data curation, and validation.Guanhao He is an Associate Professor the Department of Public Health and Preventive Medicine at Jinan University. His research interests focus environmental health and health data science. His contributions to this paper include data curation and manuscript review.Jianxiong Hu is a Postdoc Researcher at the First Affiliated Hospital of Jinan University. His research interests focus on climate change and spatiotemporal data mining. His contributions to this paper include data curation and visualization.Tao Liu is a Professor in the Department of Public Health and Preventive Medicine at Jinan University. His research interests focus on environmental health and infectious disease modeling. His contributions to this paper include conceptualization, as well as manuscript review and editing.Wenjun Ma is a Professor in the Department of Public Health and Preventive Medicine at Jinan University. His research interests focus on health data science and infectious disease modeling. His contributions to this paper include conceptualization, as well as manuscript review and editing.
2026/06/03
论文
GIScience & Remote Sensing
UrbanCompLab
GIS
城市内部土地利用多情景模拟与驱动因素分析:以中国惠城区为例
Multiple intra-urban land use simulations and driving factors analysis: A case study in Huicheng, China

城市内部土地利用变化的模拟研究逐渐受到关注,因其在决策制定与政策形成方面具有重要参考价值。尽管以往研究多集中于城市内部尺度模拟方法的开发,但针对驱动城市内部土地利用变化因素的研究仍较为匮乏。城市规划者高度关注城市内部结构的形成机制及其运行规律。为此,本文基于随机森林(RF)算法构建了元胞自动机(CA)模拟模型,旨在模拟多种城市内部土地利用变化情景,并识别不同驱动因素的贡献程度。本研究引入交通区位、环境条件、公共服务设施及人口密度等多类空间变量作为驱动因子,以深化对城市内部土地利用动态演变的理解。该模型以中国广东省惠州市惠城区2000—2010年实际历史土地利用数据进行验证,并基于验证后的模型模拟生成了2015年的多情景城市内部土地利用分布图。同时,采用随机森林算法的袋外(OOB)误差估计方法计算各空间变量的重要性指标(VIMs),进而评估并分析各驱动因素在该区域中的相对重要性。本研究为城市规划者及相关学者提供了详实、有针对性的信息,有助于制定面向不同类型城市内部土地利用的具体规划策略,并支持该地区未来的可持续发展。

Zhang, Dachuan, Liu, Xiaoping, Wu, Xiaoyu
2019/01/01
1 / 46