基于LGCP的城市管理事件空间点模式分析

doi:10.18306/dlkxjz.2020.08.010

地理科学进展 ›› 2020, Vol. 39 ›› Issue (8): 1356-1366.doi: 10.18306/dlkxjz.2020.08.010

基于LGCP的城市管理事件空间点模式分析

董文钱¹(), 董良²^,³, 向琳¹^,^*(), 陶海军¹, 赵传虎⁴, 曲寒冰²^,³

1. 中国计量大学信息工程学院,杭州 310018
2. 北京市科学技术研究院,北京 100089
3. 北京市新技术应用研究所,北京 100094
4. 河北工业大学人工智能与数据科学学院,天津 300401

收稿日期:2019-07-09 修回日期:2019-11-04 出版日期:2020-08-28 发布日期:2020-10-28
通讯作者: 向琳
作者简介:董文钱(1995— ),男,浙江温州人,硕士生,主要从事时空数据挖掘研究。E-mail: wqdong.chn@live.com
基金资助:
国家重点研发计划项目(2018YFC0809700);国家重点研发计划项目(2018YFF0301000);国家重点研发计划项目(2018YFC0704800);国家自然科学基金重大研究计划重点支持项目(91746207);北京市科学技术研究院创新团队项目(IG201801N)

Spatial pattern of urban management cases based on Log Gaussian Cox Processes

DONG Wenqian¹(), DONG Liang²^,³, XIANG Lin¹^,^*(), TAO Haijun¹, ZHAO Chuanhu⁴, QU Hanbing²^,³

1. College of Information Engineering, China Jiliang University, Hangzhou 310018, China
2. Beijing Academy of Science and Technology, Beijing 100089, China
3. Beijing Institute of New Technology Applications, Beijing 100094, China
4. School of Artificial Intelligence, Hebei University of Technology, Tianjin 300401, China

Received:2019-07-09 Revised:2019-11-04 Online:2020-08-28 Published:2020-10-28
Contact: XIANG Lin
Supported by:
National Key Research and Development Program of China(2018YFC0809700);National Key Research and Development Program of China(2018YFF0301000);National Key Research and Development Program of China(2018YFC0704800);National Natural Science Foundation of China(91746207);Innovation Team Project of Beijing Academy of Science and Technology(IG201801N)

摘要/Abstract

摘要：

数字化城管系统累积了大量城管事件历史数据,充分挖掘事件背后的空间分布模式和事件成因机制能够为城管部门的管控工作提供决策支持。论文利用Log Gaussian Cox Processes(LGCP)模型分析了西北某地H市P区的街面秩序类、市容环境类和宣传广告类城管事件之间的空间分布差异和事件成因影响差异。研究发现：① 3类城管事件都呈现出明显的空间聚集,其空间聚集尺度最远不超过924 m;② 各类事件聚集的特征各不相同,街面秩序类贴近城区主要干道,呈路网状。市容环境类表现出在城区中心块状聚集,周边地区零星分散的特征。宣传广告类靠近交通干线呈长条状,靠近商业中区域呈块状分布;③ 城区内不同类别的POI对城管事件的影响大小不同。购物服务类、医疗保健类和居民住宅类表现出最显著的影响,说明特定区域内人群的流量和密度是影响城管事件分布的重要因素,人群的流动和聚集会加剧城管事件数量的增加。研究结果能够满足城管部门的城管事件空间分布热点识别以及事件成因分析的需求。

关键词: 城管事件, Log Gaussian Cox Processes, 空间点模式, 积分嵌套拉普拉斯逼近, 随机偏微分方程

Abstract:

Digital urban management systems have accumulated a large amount of historical data of urban management cases, but there is a general lack of research on the overall spatial pattern and cause of urban management cases. Therefore, it is necessary to fully explore the spatial distribution pattern and cause mechanism behind the incidents of urban management, which could provide decision support for the urban management departments to prevent and control the cases. Taking street order, urban environment, and publicity advertising urban management cases as an example and considering the points of interest (POI) data, this study used the Log Gaussian Cox Processes (LGCP) model to analyze the differences of spatial distribution and influencing factors between street order cases, urban environment cases, and publicity advertising cases in P district of H city, Northwest China. The study found that: 1) All the three types of urban management cases present obvious spatially aggregated distribution, and no spatial correlation is believed to exist beyond 924 meters. 2) The spatial features of the agglomeration space are different. The street order cases are close to the main trunk roads in the urban area, resembling a road network. The urban environment cases tend to cluster as blocks around the center of the district, while more scattered and dispersed in the peripheral areas of the district. The publicity advertising cases are in elongated distribution near the main traffic lines, but clustered as blocks in the commercial areas of the city. 3) Different types of POI in the study area have different impacts. Shopping services, health care, and residential areas show the most significant attractiveness, indicating that the flow and density of people in specific areas are the most important factors that affect the distribution of urban management cases, and increased flow and concentration of the crowd will sharply increase the number of urban management incidents. The results of this study include spatial hotspot identification and cause analysis, which can meet the urban management departments' needs.

Key words: urban management cases, Log Gaussian Cox Processes, spatial point pattern, integrated nested Laplace approximation (INLA), stochastic partial differential equations (SPDE)

董文钱, 董良, 向琳, 陶海军, 赵传虎, 曲寒冰. 基于LGCP的城市管理事件空间点模式分析[J]. 地理科学进展, 2020, 39(8): 1356-1366.

DONG Wenqian, DONG Liang, XIANG Lin, TAO Haijun, ZHAO Chuanhu, QU Hanbing. Spatial pattern of urban management cases based on Log Gaussian Cox Processes[J]. PROGRESS IN GEOGRAPHY, 2020, 39(8): 1356-1366.

图/表 9

表1

图1

图2

表2

图3

图4

图5

表3

LGCP模型的随机场超参数后验估计值"

	参数	均值	标准差	2.5%分位数	97.5%分位数
街面秩序类	$σ S (x) 2$	1.5161	0.1219	1.2946	1.7731
街面秩序类	$ρ$	0.6897	0.0324	0.6297	0.7567
市容环境类	$σ S (x) 2$	0.9505	0.0816	0.8024	1.1227
市容环境类	$ρ$	0.7210	0.0372	0.6522	0.7983
宣传广告类	$σ S (x) 2$	1.2771	0.1230	1.0574	1.5396
宣传广告类	$ρ$	0.8279	0.0460	0.7437	0.9242

表3

表4

LGCP模型的协变量参数后验估计值"

城管事件类型	协变量	均值	标准差	2.5%分位数	97.5%分位数
街面秩序类事件	截距	0.9891	0.0122	0.9656	1.0130
	$D x 1$	0.3906^*	0.0726	0.3387	0.4503
	$D x 2$	0.8288^*	0.0526	0.7474	0.9189
	$D x 3$	1.7451^*	0.0454	1.5963	1.9075
	$D x 4$	4.9590^*	0.0848	4.1980	5.8574
	$D x 5$	1.0134	0.0828	0.8613	1.1921
	$D x 6$	1.2264^*	0.0272	1.1626	1.2936
	$D x 7$	1.1971^*	0.0444	1.0972	1.3060
	$D x 8$	0.5372^*	0.0815	0.4577	0.6304
市容环境类事件	截距	0.6619^*	0.0161	0.6414	0.6832
	$D x 1$	0.3364^*	0.0749	0.2904	0.3895
	$D x 2$	0.9001	0.0593	0.8028	1.0092
	$D x 3$	1.4776^*	0.0493	1.3411	1.6276
	$D x 4$	5.9174^*	0.0880	4.9789	7.0315
	$D x 5$	1.6981^*	0.0869	1.4318	2.0138
	$D x 6$	0.6726^*	0.0293	0.6350	0.7124
	$D x 7$	3.6503^*	0.0465	3.3318	3.9992
	$D x 8$	0.2838^*	0.0923	0.2368	0.3402
宣传广告类事件	截距	0.8779^*	0.0135	0.8550	0.9015
	$D x 1$	0.3322^*	0.0763	0.2860	0.3858
	$D x 2$	0.7966^*	0.0557	0.7141	0.8884
	$D x 3$	0.8907^*	0.0475	0.8113	0.9777
	$D x 4$	9.9115^*	0.0894	8.3170	11.8106
	$D x 5$	1.2930^*	0.0879	1.0882	1.5363
	$D x 6$	0.8584^*	0.0321	0.8059	0.9141
	$D x 7$	2.8320^*	0.0486	2.5744	3.1152
	$D x 8$	0.4776^*	0.0897	0.4004	0.5696

表4

参考文献 36

[1]	Knox P L, McCarthy L M. 城市化: 城市地理学导论 [M]. 3版. 姜付仁, 万金红, 董磊华, 等, 译. 北京: 电子工业出版社, 2016.
	[ Knox P L, McCarthy L M. Urbanization: An introduction to urban geography. 3rd Edition. Translated by Jiang Furen, Wan Jinhong, Dong Leihua, et al. Beijing, China: Publishing House of Electronics Industry, 2016. ]
[2]	李颖玥, 刘朝晖. 中国数字化城市管理发展综述[J]. 智能建筑与智慧城市, 2017(2):28-32.
	[ Li Yingyue, Liu Zhaohui. Review on the development of digital city management in China. Intelligent Building & Smart City, 2017(2):28-32. ]
[3]	徐智邦, 周亮, 蓝婷, 等. 基于POI数据的巨型城市消防站空间优化: 以北京市五环内区域为例[J]. 地理科学进展, 2018,37(4):535-546. doi: 10.18306/dlkxjz.2018.04.009
	[ Xu Zhibang, Zhou Liang, Lan Ting, et al. Spatial optimization of mega-city fire station distribution based on point of interest data: A case study within the 5th Ring Road in Beijing. Progress in Geography, 2018,37(4):535-546. ]
[4]	曾璇, 崔海山, 刘毅华. 基于网络空间点模式的餐饮店空间格局分析[J]. 地球信息科学学报, 2018,20(6):837-843.
	[ Zeng Xuan, Cui Haishan, Liu Yihua. Analysis on spatial distribution characteristics of restaurant based on network spatial point model. Journal of Geo-Information Science, 2018,20(6):837-843. ]
[5]	李钢, 陈未雨, 杨兰, 等. 武汉市快递自提点的空间格局与集聚模式研究[J]. 地理科学进展, 2019,38(3):407-416.
	[ Li Gang, Chen Weiyu, Yang Lan, et al. Spatial pattern and agglomeration mode of parcel collection and delivery points in Wuhan City. Progress in Geography, 2019,38(3):407-416. ]
[6]	柳林, 姜超, 周素红, 等. 城市入室盗窃犯罪的多尺度时空格局分析: 基于中国H市DP半岛的案例研究[J]. 地理研究, 2017,36(12):2451-2464.
	[ Liu Lin, Jiang Chao, Zhou Suhong, et al. Spatial-temporal patterns of burglary at multiple scales: The case of DP Peninsula in H City, China. Geographical Research, 2017,36(12):2451-2464. ]
[7]	强思维, 陈夏明, 姜开达, 等. 基于移动网络流量日志的城市时空行为分析[J]. 计算机研究与发展, 2016,53(4):932-940.
	[ Qiang Siwei, Chen Xiaming, Jiang Kaida, et al. Urban spatio-temporal behavior analysis based on mobile network traffic logs. Journal of Computer Research and Development, 2016,53(4):932-940. ]
[8]	袁坚, 王鹏, 王钺, 等. 基于时空特征的城市轨道交通客流量预测方法[J]. 北京交通大学学报, 2017,41(6):42-48.
	[ Yuan Jian, Wang Peng, Wang Yue, et al. A passenger volume prediction method based on temporal and spatial characteristics for urban rail transit. Journal of Beijing Jiaotong University, 2017,41(6):42-48. ]
[9]	赵永. 空间数据统计分析的思想起源与应用演化[J]. 地理研究, 2018,37(10):2058-2074.
	[ Zhao Yong. The origin and application history of statistical analysis of spatial data. Geographical Research, 2018,37(10):2058-2074. ]
[10]	佘冰, 朱欣焰, 呙维, 等. 基于空间点模式分析的城市管理事件空间分布及演化: 以武汉市江汉区为例[J]. 地理科学进展, 2013,32(6):924-931.
	[ She Bing, Zhu Xinyan, Guo Wei, et al. Spatial distribution and evolution of city management events based on the spatial point pattern analysis: A case study of Jianghan District, Wuhan City. Progress in Geography, 2013,32(6):924-931. ]
[11]	伏家云, 靖常峰, 杜明义, 等. 参数优化DBSCAN算法的城管案件聚类分析[J]. 测绘科学, 2018,43(8):135-140.
	[ Fu Jiayun, Jing Changfeng, Du Mingyi, et al. Urban management case spatial clustering analysis based on a parameter optimized DBSCAN algorithm. Science of Surveying and Mapping, 2018,43(8):135-140. ]
[12]	靖常峰, 付艳丽, 戴培培, 等. 近邻指数的城管案件全局分布模式分析[J]. 测绘科学, 2017,42(6):76-81.
	[ Jing Changfeng, Fu Yanli, Dai Peipei, et al. Spatial pattern analysis on global distribution of urban management case based on nearest index. Science of Surveying and Mapping, 2017,42(6):76-81. ]
[13]	聂可, 王振声, 杜清运, 等. 网络约束点格局分析研究综述[J]. 地理科学进展, 2017,36(10):1196-1207.
	[ Nie Ke, Wang Zhensheng, Du Qingyun, et al. A review of network-constrained point pattern analysis. Progress in Geography, 2017, 36(10):1196-1207. ].
[14]	马志波, 肖文发, 黄清麟, 等. 生态学中的点格局研究概况及其在国内的应用[J]. 生态学报, 2017,37(19):6624-6632.
	[ Ma Zhibo, Xiao Wenfa, Huang Qinglin, et al. A review of point pattern analysis in ecology and its application in China. Acta Ecologica Sinica, 2017,37(19):6624-6632. ]
[15]	Illian J, Penttinen A, Stoyan H, et al. Statistical analysis and modelling of spatial point patterns[M]. London, UK: John Wiley and Sons, 2008: 110-123.
[16]	裴韬, 李婷, 周成虎. 时空点过程: 一种新的地学数据模型、分析方法和观察视角[J]. 地球信息科学学报, 2013,15(6):793-800.
	[ Pei Tao, Li Ting, Zhou Chenghu. Spatiotemporal point process: A new data model, analysis methodology and viewpoint for geoscientific problem. Journal of Geo-information Science, 2013,15(6):793-800. ]
[17]	Illian J B, Martino S, Sørbye S H, et al. Fitting complex ecological point process models with integrated nested Laplace approximation[J]. Methods in Ecology and Evolution, 2013,4(4):305-315.
[18]	Waagepetersen R, Guan Y, Jalilian A, et al. Analysis of multispecies pointpatterns by using multivariate Log-Gaussian Cox Processes[J]. Journal of the Royal Statistical Society: Series C (Applied Statistics), 2016,65(1):77-96.
[19]	Law R, Illian J, Burslem D F, et al. Ecological information from spatial patterns of plants: Insights from point process theory[J]. Journal of Ecology, 2009,97(4):616-628.
[20]	Wiegand T, Moloney K A. Handbook of spatial point-pattern analysis in ecology [M]. Boca Raton, USA: CRC Press, 2013: 26-39.
[21]	Velázquez E, Martínez I, Getzin S, et al. An evaluation of the state of spatial point pattern analysis in ecology[J]. Ecography, 2016,39(11):1042-1055.
[22]	Møller J, Syversveen A R, Waagepetersen R P. Log Gaussian Cox Processes[J]. Scandinavian Journal of Statistics, 1998,25(3):451-482.
[23]	Diggle P J. Statistical analysis of spatial and spatio-temporal point patterns, third edition [M]. Boca Raton, USA: CRC Press, 2013: 167-172.
[24]	Dong G, Ma J, Wei R, et al. Electric vehicle charging point placement optimisation by exploiting spatial statistics and maximal coverage location models[J]. Transportation Research Part D: Transport and Environment, 2019,67:77-88.
[25]	Cui Q, Wang N, Haenggi M. Spatial point process modeling of vehicles in large and small cities [R]. GLOBECOM 2017-2017 IEEE Global Communications Conference. Singapore, Singapore: IEEE, 2017: 1-7.
[26]	Møller J, Waagepetersen R P. Modern statistics for spatial point processes[J]. Scandinavian Journal of Statistics, 2007,34(4):643-684.
[27]	Diggle P J, Moraga P, Rowlingson B, et al. Spatial and spatio-temporal Log-Gaussian Cox Processes: Extending the geostatistical paradigm[J]. Statistical Science, 2013,28(4):542-563.
[28]	Taylor B, Davies T, Rowlingson B, et al. Bayesian inference and data augmentation schemes for spatial, spatiotemporal and multivariate Log-Gaussian Cox Processes in R[J]. Journal of Statistical Software, 2015,63(7):1-48.
[29]	Lindgren F, Rue H, Lindström J. An explicit link between Gaussian fields and Gaussian Markov random fields: The stochastic partial differential equation approach[J]. Journal of the Royal Statistical Society: Series B (Statistical Methodology), 2011,73(4):423-498.
[30]	Illian J B, Sørbye S H, Rue H. A toolbox for fitting complex spatial point process models using integrated nested Laplace approximation (INLA)[J]. The Annals of Applied Statistics, 2012,6(4):1499-1530.
[31]	Simpson D, Illian J B, Lindgren F, et al. Going off grid: Computationally efficient inference for Log-Gaussian Cox Processes[J]. Biometrika, 2016,103(1):49-70.
[32]	Cameletti M, Lindgren F, Simpson D, et al. Spatio-temporal modeling of particulate matter concentration through the SPDE approach[J]. Advances in Statistical Analysis, 2013,97(2):109-131.
[33]	Blangiardo M, Cameletti M. Spatial and spatio-temporal Bayesian models with R-INLA[M]. London, UK: John Wiley and Sons, 2015: 193-198.
[34]	Teng M, Nathoo F, Johnson T D. Bayesian computation for Log-Gaussian Cox Processes: A comparative analysis of methods[J]. Journal of Statistical Computation and Simulation, 2017,87(11):2227-2252. doi: 10.1080/00949655.2017.1326117 pmid: 29200537
[35]	Rue H, Martino S, Chopin N. Approximate Bayesian inference for latent Gaussian models by using integrated nested Laplace approximations[J]. Journal of the Royal Statistical Society: Series B (Statistical Methodology), 2009,71(2):319-392.
[36]	Zhang S, Liu X, Tang J, et al. Spatio-temporal modeling of destination choice behavior through the Bayesian hierarchical approach[J]. Physica A: Statistical Mechanics and Its Applications, 2018,512:537-551.

事件类别	事件内容	事件数量/件
街面秩序	乱堆物堆料、占道经营、流动摊贩、机动车乱停放等	27016
市容环境	暴露垃圾、道路不洁、绿化不洁、积存垃圾渣土等	12429
宣传广告	散发小广告、违章悬挂物品、占道广告牌等	20434

POI类别	POI内容	POI数量
餐饮服务类	中餐厅、快餐厅、外国餐厅、休闲餐饮场所等	2096
交通设施类	公交车站、物流站等	138
居民住宅类	住宅区	292
购物服务类	便利店、超市、烟酒专卖店、购物相关场所等	1073
生活服务类	美容美发店、洗浴推拿场所、生活服务场所等	1103
住宿服务类	酒店、旅馆、招待所等	113
医疗保健类	医院、社区诊所、卫生院、药店、药房等	385
公司企业类	公司、企业、工厂等	773

基于LGCP的城市管理事件空间点模式分析

Spatial pattern of urban management cases based on Log Gaussian Cox Processes

RichHTML

PDF (PC)

赞

可视化

摘要/Abstract

引用本文

使用本文

图/表 9

参考文献 36

相关文章 1

Metrics

本文评价

推荐阅读 0