地球信息科学学报 ›› 2023, Vol. 25 ›› Issue (10): 1986-1999.doi: 10.12082/dqxxkx.2023.230299
贺日兴1,2(), 唐宗棣1,2, 姜超3, 林艳4, 陆宇梅1,2, 李欣然1,2, 龙伟1,2, 邓悦1,2,*(
)
收稿日期:
2023-05-31
修回日期:
2023-08-02
出版日期:
2023-10-25
发布日期:
2023-09-22
通讯作者:
* 邓悦(1994—),女,新疆哈巴河人,博士,讲师,主要研究方向为空间数据挖掘、空间分析。 E-mail: 6921@cnu.edu.cn作者简介:
贺日兴(1972—),男,江西安福人,博士,教授,博导,主要从事犯罪地理和警用地理信息技术等研究。E-mail: herixing@cnu.edu.cn
基金资助:
HE Rixing1,2(), TANG Zongdi1,2, JIANG Chao3, LIN Yan4, LU Yumei1,2, LI Xinran1,2, LONG Wei1,2, DENG Yue1,2,*(
)
Received:
2023-05-31
Revised:
2023-08-02
Online:
2023-10-25
Published:
2023-09-22
Contact:
* DENG Yue, E-mail: Supported by:
摘要:
传统的犯罪地理和犯罪时空预测方法主要是以警务辖区或格网为基本单元,分析结果不利于指导精细化的巡防警力规划部署。基于深度学习的图神经网络方法可以自然地与微观尺度下的路网拓扑结构相结合,实现道路尺度下的精细犯罪预测,但现有方法鲜有考虑道路权重对预测结果的影响。本文通过引入道路通达度和距离衰减因子,构建了一种顾及道路权重的图卷积犯罪时空预测模型(Road Weighted Spatio-Temporal Graph Convolutional Network,RW-STGCN),并利用芝加哥2016—2017年街面盗窃犯罪数据对模型进行评估。结果表明: ① 与未考虑道路权重的时空图卷积模型相比,RW-STGCN模型命中率在不同的路网覆盖比例下(1%、5%、10%、20%)的提升均在6.5%以上,且随着覆盖比例的下降,模型命中率的提升更为显著,最大提升超过了50%; ② 模型消融性实验表明,同时考虑2种道路权重的模型比仅考虑距离衰减权重或道路通达度权重单个因子的模型预测性能提升更为明显,命中率最大提升了12.9%。本研究构建的RW-STGCN模型有助于街面类犯罪预测,可为警务部门基于路网进行科学巡逻防控规划与警力部署提供辅助决策支持,此外还可用于以道路作为分析单元的城市计算问题研究。
贺日兴, 唐宗棣, 姜超, 林艳, 陆宇梅, 李欣然, 龙伟, 邓悦. 顾及道路权重的图卷积犯罪时空预测模型[J]. 地球信息科学学报, 2023, 25(10): 1986-1999.DOI:10.12082/dqxxkx.2023.230299
HE Rixing, TANG Zongdi, JIANG Chao, LIN Yan, LU Yumei, LI Xinran, LONG Wei, DENG Yue. A Graph Convolution-based Spatio-temporal Crime Prediction Model Considering Road Weights[J]. Journal of Geo-information Science, 2023, 25(10): 1986-1999.DOI:10.12082/dqxxkx.2023.230299
[1] | 王发曾. 我国城市犯罪空间防控研究二十年[J]. 人文地理, 2010, 25(4):25-30. |
[Wang F Z. A review of urban spatial anti-crime study in China from 1980 to 2000[J]. Human Geography, 2010, 25(4):25-30.] DOI:10.13959/j.issn.1003-2398.2010.04.011 | |
[2] |
贺日兴, 陆宇梅, 姜超, 等. 近10年来犯罪时空预测国内外研究与实践进展[J]. 地球信息科学学报, 2023, 25(4):866-882.
doi: 10.12082/dqxxkx.2023.220808 |
[He R X, Lu Y M, Jiang C, et al. Progress in research and practice of spatial-temporal crime prediction over the past decade[J]. Journal of Geo-information Science, 2023, 25(4):866-882.] DOI:10.12082/dqxxkx.2023.220808 | |
[3] | Perry W L, Mcinnis B, Price C C, et al. Predictive policing: the role of crime forecasting in law enforcement operations[M]. Santa Monica, CA: RAND Corporation, 2013. |
[4] | Townsley M, Homel R, Chaseling J. Infectious burglaries. A test of the near repeat hypothesis[J]. The British Journal of Criminology, 2003, 43(3):615-633. DOI:10.1093/bjc/43.3.615 |
[5] | Kafadar K, Bowman A W, Azzalini A. Applied smoothing techniques for data analysis: the kernel approach with S-PLUS illustrations[J]. Journal of the American Statistical Association, 1999, 94(447):982. DOI:10.2307/2670015 |
[6] | 贺日兴. 警用地理信息系统——理论、技术与应用[M]. 北京: 中国人民公安大学出版社, 2022:240-248. |
[He R X. Police Geographic Information System - Theory, Technology, and Application[M]. Beijing: China Renmin University Press, 2022:240-248.] | |
[7] | Kennedy L W, Caplan J M, Piza E. Risk clusters, hotspots, and spatial intelligence: Risk terrain modeling as an algorithm for police resource allocation strategies[J]. Journal of Quantitative Criminology, 2011, 27(3):339-362. DOI:10.1007/s10940-010-9126-2 |
[8] | Mohler G O, Short M B, Brantingham P J, et al. Self-exciting point process modeling of crime[J]. Journal of the American Statistical Association, 2011, 106(493):100-108. DOI:10.1198/jasa.2011.ap09546 |
[9] | 李卫红, 闻磊, 陈业滨. 改进的GA-BP神经网络模型在财产犯罪预测中的应用[J]. 武汉大学学报·信息科学版, 2017, 42(8):1110-1116,1171. |
[Li W H, Wen L, Chen Y B. Property crime forecast based on improved GA-BP neural network model[J]. Geomatics and Information Science of Wuhan University, 2017, 42(8):1110-1116,1171.] DOI:10.13203/j.whugis20160911 | |
[10] |
柳林, 刘文娟, 廖薇薇, 等. 基于随机森林和时空核密度方法的不同周期犯罪热点预测对比[J]. 地理科学进展, 2018, 37(6):761-771.
doi: 10.18306/dlkxjz.2018.06.003 |
[Liu L, Liu W J, Liao W W, et al. Comparison of random forest algorithm and space-time kernel density mapping for crime hotspot prediction[J]. Progress in Geography, 2018, 37(6):761-771.] DOI:10.18306/dlkxjz.2018.06.003 | |
[11] | Hardyns W, Rummens A. Predictive policing as a new tool for law enforcement? recent developments and challenges[J]. European Journal on Criminal Policy and Research, 2018, 24:201-218. DOI:10.1007/s10610-017-9361-2 |
[12] | Baculo M J C, Marzan C S, de Dios Bulos R, et al. Geospatial-temporal analysis and classification of criminal data in Manila[C]//2017 2nd IEEE International Conference on Computational Intelligence and Applications (ICCIA).IEEE, 2017:6-11. DOI:10.1109/CIAPP.2017.8167050 |
[13] | 刘大千. 长春市犯罪空间分析及规划管理防控[D]. 长春: 东北师范大学, 2012. |
[Liu D Q. Spatial analysis of crimes in Changchun and crime control from planning and administrative perspective[D]. Changchun: Northeast Normal University, 2012.] | |
[14] | Han X G, Hu X F, Wu H G, et al. Risk prediction of theft crimes in urban communities: An integrated model of LSTM and ST-GCN[J]. IEEE Access, 2020, 8:217222-217230. DOI:10.1109/ACCESS.2020.3041924 |
[15] | 陈鹏, 林平, 孙菲菲, 等. 风险地形建模在犯罪风险评估中的应用[J]. 测绘与空间地理信息, 2017, 40(12):4-9. |
[Chen P, Lin P, Sun F F, et al. The application of risk terrain model in risk assessment of crime[J]. Geomatics & Spatial Information Technology, 2017, 40(12):4-9.] DOI:10.3969/j.issn.1672-5867.2017.12.002 | |
[16] |
徐冲, 柳林, 周素红. 基于临近相似性考虑的犯罪热点密度图预测准确性比较——以DP半岛街头抢劫犯罪为例[J]. 地理科学, 2016, 36(1):55-62.
doi: 10.13249/j.cnki.sgs.2016.01.007 |
[Xu C, Liu L, Zhou S H. The comparison of predictive accuracy of crime hotspot density maps with the consideration of the near similarity: A case study of robberies at DP peninsula[J]. Scientia Geographica Sinica, 2016, 36(1):55-62.] DOI:10.13249/j.cnki.sgs.2016.01.007 | |
[17] | 毛媛媛, 戴慎志. 犯罪空间分布与环境特征——以上海市为例[J]. 城市规划学刊, 2006(3):85-93. |
[Mao Y Y, Dai S Z. Research on spatial and environmental characters of crimes: Case study of Shanghai[J]. Urban Planning Forum, 2006(3):85-93.] DOI:10.3969/j.issn.1000-3363.2006.03.013 | |
[18] |
宋广文, 黎晓彐, 肖露子, 等. 交互作用视角下人口流动性与住房类型对城市入室盗窃空间格局的影响[J]. 地理研究, 2022, 41(11):2897-2911.
doi: 10.11821/dlyj020220350 |
[Song G W, Li X X, Xiao L Z, et al. Effects of population mobility and housing type on spatial pattern of urban burglary from the perspective of interaction[J]. Geographical Research, 2022, 41(11):2897-2911.] DOI:10.11821/dlyj020220350 | |
[19] |
张延吉, 林钦熙, 朱春武, 等. 可渗透性环境对盗窃犯罪分布的影响及社会解组的调节作用——兼论街道眼理论与防卫空间理论的适用性[J]. 地理科学进展, 2022, 41(6):1041-1052.
doi: 10.18306/dlkxjz.2022.06.008 |
[Zhang Y J, Lin Q X, Zhu C W, et al. The influence of permeable built environment on theft crime pattern and the moderation effects of social disorganization: Applicability of street eyes and defensible place theories[J]. Progress in Geography, 2022, 41(6):1041-1052.] DOI:10.18306/dlkxjz.2022.06.008 | |
[20] | 柳林, 杜方叶, 肖露子, 等. 不同类型道路密度对公共空间盗窃犯罪率的影响——基于ZG市的实证研究[J]. 人文地理, 2017, 32(6):32-38, 46. |
[Liu L, Du F Y, Xiao L Z, et al. The density of various road types and larceny rate: An empirical analysis of zg city[J]. Human Geography, 2017, 32(6):32-38,46.] DOI:10.13959/j.issn.1003-2398.2017.06.004 | |
[21] | Sadeek S N, Minhuz Uddin Ahmed A J M, Hossain M, et al. Effect of land use on crime considering exposure and accessibility[J]. Habitat International, 2019, 89:102003. DOI:10.1016/j.habitatint.2019.102003 |
[22] |
Rosser G, Davies T P, Bowers K J, et al. Predictive crime mapping: arbitrary grids or street networks?[J]. Journal of Quantitative Criminology, 2017, 33(3):569-594. DOI:10.1007/s10940-016-9321-x
pmid: 32025086 |
[23] | Okabe A, Satoh T, Sugihara K. A kernel density estimation method for networks, its computational method and a GIS-based tool[J]. International Journal of Geographical Information Science, 2009, 23(1):7-32. DOI:10.1080/13658810802475491 |
[24] | Shiode S, Shiode N. Microscale prediction of near-future crime concentrations with street-level geosurveillance[J]. Geographical Analysis, 2014, 46(4):435-455. DOI:10.1111/gean.12065 |
[25] | Zhou B B, Chen L B, Zhou F X, et al. Dynamic road crime risk prediction with urban open data[J]. Frontiers of Computer Science, 2022, 16(1):161609. DOI:10.1007/s11704-021-0136-z |
[26] | Zhang Y, Cheng T. Graph deep learning model for network-based predictive hotspot mapping of sparse spatio-temporal events[J]. Computers, Environment and Urban Systems, 2020, 79:101403. DOI:10.1016/j.compenvurbsys.2019.101403 |
[27] | Yu B, Yin H T, Zhu Z X. Spatio-temporal graph convolutional networks: A deep learning framework for traffic forecasting[EB/OL]. 2017: arXiv:1709.04875. https://arxiv.org/abs/1709.04875 |
[28] | Sun J K, Zhang J B, Li Q F, et al. Predicting citywide crowd flows in irregular regions using multi-view graph convolutional networks[J]. IEEE Transactions on Knowledge and Data Engineering, 2022, 34(5):2348-2359. DOI:10.1109/TKDE.2020.3008774 |
[29] | Zhao L, Song Y J, Zhang C, et al. T-GCN: A temporal graph convolutional network for traffic prediction[J]. IEEE Transactions On Intelligent Transportation Systems, 2020, 21(9):3848-3858. DOI:10.1109/TITS.2019.2 935152 |
[30] | Veličković P, Cucurull G, Casanova A, et al. Graph attention networks[EB/OL]. 2017: arXiv:1710.10903. https://arxiv.org/abs/1710.10903 |
[31] | Guo S N, Lin Y F, Feng N, et al. Attention based spatial-temporal graph convolutional networks for traffic flow forecasting[J]. Proceedings of the AAAi Conference on Artificial Intelligence, 2019, 33(1):922-929. DOI:10.1609/aaai.v33i01.3301922 |
[32] | Hamilton W L, Ying R, Leskovec J. Inductive representation learning on large graphs[EB/OL]. 2017:arXiv:1706.02216. https://arxiv.org/abs/1710.10903. |
[33] | Johnson S D, Bowers K J, Birks D, et al. Burglary prediction, theory, flow and friction[M]. CRC Press, 2005:203-223. |
[34] | Johnson S D, Bowers K J. Near repeats and crime forecasting[M]. Encyclopedia of Criminology and Criminal Justice, Bruinsma G, Weisburd D, New York, NY: Springer New York, 2014:3242-3254. |
[35] |
肖露子, 柳林, 宋广文, 等. 基于理性选择理论的社区环境对入室盗窃的影响研究[J]. 地理研究, 2017, 36(12):2479-2491.
doi: 10.11821/dlyj201712017 |
[Xiao L Z, Liu L, Song G W, et al. Impacts of community environment on residential burglary based on rational choice theory[J]. Geographical Research, 2017, 36(12):2479-2491.] DOI:10.11821/dlyj201712017 | |
[36] | Dong H W. Does walkability undermine neighbourhood safety?[J]. Journal of Urban Design, 2017, 22(1):59-75. DOI:10.1080/13574809.2016.1247644 |
[37] | Dauphin Y N, Fan A, Auli M, et al. Language modeling with gated convolutional networks[EB/OL]. 2016: arXiv: 1612.08083. https://arxiv.org/abs/1612.08083 |
[38] | Zaremba W, Sutskever I, Vinyals O. Recurrent neural network regularization[EB/OL]. 2014: arXiv:1409.2329. https://arxiv.org/abs/1409.2329 |
[39] | Shi X J, Chen Z R, Wang H, et al. Convolutional LSTM network: A machine learning approach for precipitation nowcasting[EB/OL]. 2015:arXiv:1506.04214. https://arxiv.org/abs/1506.04214 |
[40] | 谢超. 基于网络分析的地下通道对街区步行可达性影响研究——以广州天河新城为例[C].成都://面向高质量发展的空间治理——2021中国城市规划年会论文集(05城市规划新技术应用), 2021:444-453. |
[Xie C. Research on the impact of underground passages on pedestrian accessibility of street blocks based on network analysis: A case study of Tianhe new city in Guangzhou[C]. Chengdu Sichuan//Spatial governance for high-quality development: Proceedings of the 2021 China Urban Planning Annual Conference (05 Urban Planning New Technology Application), 2021:444-453.] | |
[41] |
陈方, 张澄洋, 丁思远. 生活街区轨道站步行易达性测度研究——以深圳和香港为例[J]. 南方建筑, 2021(4):31-38.
doi: 10.3969/j.issn.1000-0232.2021.04.031 |
[Chen F, Zhang C Y, Ding S Y. A study on the measurement of pedestrian accessibility of living-block rail transit station: A case study of Shenzhen and Hong Kong[J]. South Architecture, 2021(4):31-38.] DOI:10.3969/j.issn.1000-0232.2021.04.031 | |
[42] |
张春霞, 周素红, 柳林, 等. 建成环境对星级酒店内被盗的影响——以ZG市中心城区为例[J]. 地理科学进展, 2020, 39(5):829-840.
doi: 10.18306/dlkxjz.2020.05.011 |
[Zhang C X, Zhou S H, Liu L, et al. Relationship between the built environment and theft cases in star hotels in ZG central city[J]. Progress in Geography, 2020, 39(5):829-840.] DOI:10.18306/dlkxjz.2020.05.011 | |
[43] | 何健, 黄启乐, 汪振东. 广州市城市道路网密度提升路径探索[J]. 交通与运输, 2019, 35(4):35-38. |
[He J, Huang Q L, Wang Z D. Path of urban road network density enhancement in Guangzhou[J]. Traffic & Transportation, 2019, 35(4):35-38.] DOI:10.3969/j.issn.1671-3400.2019.04.009 | |
[44] | 居少捷, 孙琳珊. 基于空间句法的轨交站点人流空间分布研究——以南京新街口为例[J]. 城市建筑, 2023, 20(6):87-89,93. |
[Ju S J, Sun L S. Study on spatial distribution of population flow in rail transit station based on space syntax: A case of Nanjing Xinjiekou[J]. Urbanism and Architecture, 2023, 20(6):87-89,93.] DOI:10.19892/j.cnki.csjz.2023.06.24 | |
[45] | 姜昀呈, 孙立坚, 王涛涛, 等. 兰州市中心城区犯罪分布与城市环境的关系[J]. 测绘科学, 2021, 46(5):167-174. |
[Jiang Y C, Sun L J, Wang T T, et al. The coupling relationship between crime distribution and urban environment in Lanzhou city center[J]. Science of Surveying and Mapping, 2021, 46(5):167-174.] DOI:10.16251/j.cnki.1009-2307.2021.05.023 | |
[46] | Farrell G, Phillips C, Pease K. LIKE TAKING CANDY: Why does repeat victimization occur?[J]. The British Journal of Criminology, 1995, 35(3):384-399. DOI:10.1093/oxfordjournals.bjc.a048523 |
[47] | Bowers K J, Johnson S D, Pease K. Prospective hot-spotting: The future of crime mapping?[J]. The British Journal of Criminology, 2004, 44:641-658. DOI:10.1093/bjc/azh036 |
[48] | Hu Y J, Wang F H, Guin C, et al. A spatio-temporal kernel density estimation framework for predictive crime hotspot mapping and evaluation[J]. Applied Geography, 2018, 99:89-97. DOI:10.1016/j.apgeog.2018.08.001 |
[49] | Song J, Son J, Seo D H, et al. ST-GAT: a spatio-temporal graph attention network for accurate traffic speed prediction[C].//Proceedings of the 31st ACM International Conference on Information & Knowledge Management New York, NY, USA: ACM, 2022:4500-4504. DOI:10.1145/3511808.3557705 |
[1] | 令振飞, 刘涛, 杜萍, 赵丹, 陈朴一, 马天恩. 一种支持建筑群组相似模式检索的变分图卷积自编码模型[J]. 地球信息科学学报, 2023, 25(7): 1405-1417. |
[2] | 贺日兴, 陆宇梅, 姜超, 邓悦, 李欣然, 时东. 近10年来犯罪时空预测国内外研究与实践进展[J]. 地球信息科学学报, 2023, 25(4): 866-882. |
[3] | 黄昕, 毛政元. 基于时空多图卷积网络的网约车乘客需求预测[J]. 地球信息科学学报, 2023, 25(2): 311-323. |
[4] | 朱佳波, 王益, 张拾斤, 陈家亮, 杨璐. 基于路径模拟和空间句法的封闭式非门禁居住小区犯罪人逃逸行为研究[J]. 地球信息科学学报, 2023, 25(11): 2178-2190. |
[5] | 张旭, 柳林, 周翰林, 岳瀚, 孙秋远. 基于贝叶斯逻辑回归模型研究百度街景图像微观建成环境因素对街面犯罪的影响[J]. 地球信息科学学报, 2022, 24(8): 1488-1501. |
[6] | 沙恒宇, 金广垠, 程光权, 黄金才, 吴克宇. 顾及地理-语义动态的城市热点预测框架[J]. 地球信息科学学报, 2022, 24(1): 25-37. |
[7] | 令振飞, 刘涛, 杜萍, 张耀蓉, 杨国林, 锁旭宏. 样本数量不平衡下的建筑群模式识别方法研究[J]. 地球信息科学学报, 2022, 24(1): 63-73. |
[8] | 王海起, 孔浩然, 李学伟. 基于过滤文本和社交网络的用户常驻位置预测[J]. 地球信息科学学报, 2021, 23(10): 1778-1786. |
[9] | 顾海硕, 陈鹏, 李慧波. 犯罪时空预测方法研究综述与展望[J]. 地球信息科学学报, 2021, 23(1): 43-57. |
[10] | 杨建思, 柳帅, 王艳东, 廖明生. 融合多源大数据的武汉城市空置区域评估与分析[J]. 地球信息科学学报, 2020, 22(5): 997-1007. |
[11] | 邓敏, 蔡建南, 杨文涛, 唐建波, 杨学习, 刘启亮, 石岩. 多模态地理大数据时空分析方法[J]. 地球信息科学学报, 2020, 22(1): 41-56. |
[12] | 陈能成, 肖长江, 杨超, 王伟. 地理空间传感网融合服务技术与应用[J]. 地球信息科学学报, 2020, 22(1): 11-20. |
[13] | 段炼, 党兰学, 胡涛, 朱欣焰, 叶信岳. 融合历史犯罪数据的疑犯社会活动位置预测[J]. 地球信息科学学报, 2018, 20(7): 929-938. |
[14] | 程昌秀, 张文尝, 陈洁, 蔡俊. 基于空间句法的地铁可达性评价分析——以2008年北京地铁规划图为例[J]. 地球信息科学学报, 2007, 9(6): 31-35. |
|