| 163 | 0 | 62 |
| 下载次数 | 被引频次 | 阅读次数 |
犯罪时空预测是预测性警务的重要工作，借助地理信息系统（GIS）技术引入细粒度特征，可实现数据扩增，从而有效提升风险预警的准确率。基于日常活动理论，结合天气、交通密度、兴趣点（POI）分布及监管力量间距等多维特征，构建多特征预测模型，并采用七种机器学习方法进行评估。结果表明，长短期记忆（LSTM）模型在时空序列预测中表现最为优异，在百米网格尺寸下预测精度评估指数（PAI）值达到205.031。多特征模型较单变量模型在犯罪预测能力上有显著提升，可多预测约17%的犯罪总数，为大数据时代下警务决策提供了新的理论依据与实践路径。
Abstract：Spatio-temporal crime prediction, a key branch of predictive policing, enhances risk warning accuracy by leveraging GIS technology for fine-grained feature extraction and data augmentation. Based on routine activity theory, this study develops a multi-feature prediction model that integrates multiple features such as weather, traffic density, POI distribution, and police patrol density. The model is evaluated using seven machine learning methods. The results indicate that the LSTM model outperforms others in spatiotemporal sequence prediction, achieving a PAI value of 205. 031 at a 100-meter grid scale. Compared to univariate models, the multivariate one significantly improves crime prediction capability, with approximately 17% more of the total crimes. These findings demonstrate the potential of GIS-based fine-grained features and multivariate modeling to improve crime prediction accuracy, providing new theoretical and practical path for police decision-making in the era of big data.
[1]叶琼元，夏一雪.情报主导警务模式的系统演化机理与发展路径[J].中国人民警察大学学报，2022,38(12):26-33.
[2]PERRY W L. Predictive policing:the role of crime forecasting in law enforcement operations[M]. Santa Monica:Rand Corporation, 2013.
[3]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.
[4]LI S, ZHUANG J, SHEN S. Dynamic forecasting conditional probability of bombing attacks based on timeseries and intervention analysis[J]. Risk analysis, 2017,37(7):1287-1297.
[5]MONROE B L. The five vs of big data political science introduction to the virtual issue on big data in political science political analysis[J]. Political analysis, 2013, 21(V5):1-9.
[6]陈冲，庞珣.非洲恐怖袭击时空规律的大数据分析：基于GIS技术和分离总体持续期模型[J].外交评论（外交学院学报），2020,37(2):121-154.
[7]GALLISON J K, ANDRESEN M A. Crime and public transportation:a case study of ottawa’s o-train system[J].Canadian journal of criminology and criminal justice,2017, 59(1):94-122.
[8]WAND M P, JONES M C. Comparison of smoothing parameterizations in bivariate kernel density estimation[J].Journal of the american statistical association, 1993, 88(422):520-528.
[9]李继萌，吴绍忠.“一带一路”沿线国家遭受恐怖袭击时空演变特征及对策[J].中国人民警察大学学报，2022,38(8):33-39.
[10]WANG B, YIN P, BERTOZZI A L, et al. Deep learning for real-time crime forecasting and its ternarization[J].Chinese annals of mathematics, series B, 2019, 40(6):949-966.
[11]CHEONG J S. Detecting crime hot spots using gam and local moran’s i[J]. International journal of contents,2012, 8(2):89-96.
[12]OPENSHAW S, CHARLTON M, WYMER C, et al. A mark 1 geographical analysis machine for the automated analysis of point data sets[J]. International journal of geographical information system, 1987, 1(4):335-358.
[13]ANSELIN L. Local indicators of spatial association—LISA[J]. Geographical analysis, 1995, 27(2):93-115.
[14]CAPLAN J M, KENNEDY L W, PIZA E L, et al. Using vulnerability and exposure to improve robbery prediction and target area selection[J]. Applied spatial analysis and policy, 2020(13):113-136.
[15]杨学锋.热点警务的犯罪学理论基础及实践评估[J].中国人民公安大学学报（社会科学版），2018,34(3):33-39.
[16]林晞楠.日常活动理论视野中我国贪污犯罪行为及防控模式研究[D].北京：中国人民公安大学，2020.
[17]BRANTINGHAM P L, BRANTINGHAM P J. 21 Nodes,paths and edges:considerations on the complexity of crime and the physical environment[M]. London:Routledge,2017.
[18]UN-HABITAT. Enhancing urban safety and security:global report on human settlements 2007[M]. London:Routledge, 2012.
[19]DING N, ZHAI Y. Crime prevention of bus pickpocketing in Beijing, China:does air quality affect crime?[J].Security journal, 2021, 34(2):262-277.
[20]SORG E T, TAYLOR R B. Community-level impacts of temperature on urban street robbery[J]. Journal of criminal justice, 2011, 39(6):463-470.
[21]KE G, MENG Q, FINLEY T, et al. LightGBM:a highly efficient gradient boosting decision tree[J]. Advances in neural information processing systems, 2017, 30:3146-3154.
[22]KIANMEHR K, ALHAJJ R. Effectiveness of support vector machine for crime hot-spots prediction[J]. Applied artificial intelligence, 2008, 22(5):433-458.
[23]ZHANG Y, CAI L, SONG G, et al. The long-term theft prediction in Beijing using machine learning algorithms:comparison and interpretation[J]. Crime&delinquency,2023:00111287231180102.
[24]HAN X, HU X, WU H, et al. Risk prediction of theft crimes in urban communities:an integrated model of LSTM and ST-GCN[J]. Ieee access, 2020(8):217222-30.
[25]BUTT U M, LETCHMUNAN S, ALI M, et al. Start:a spatiotemporal autoregressive transformer for enhancing crime prediction accuracy[J]. IEEE transactions on computational social systems, 2025, 12(6):4650-4664.
[26]BOGOMOLOV A, LEPRI B, STAIANO J, et al. Once upon a crime:towards crime prediction from demographics and mobile data[C]//Proceedings of the Proceedings of the16th International Conference on Multimodal Interaction,F, 2014.
[27]HOCHREITER S, SCHMIDHUBER J. Long short-term memory[J]. Neural computation, 1997, 9(8):1735-1780.
[28]MA X, TAO Z, WANG Y, et al. Long short-term memory neural network for traffic speed prediction using remote microwave sensor data[J]. Transportation research part c:emerging technologies, 2015(54):187-197.
[29]ARTHUR D, VASSILVITSKII S. K-Means++:the advantages of careful seeding[R]. Stanford, 2006.
[30]YU B, YIN H, ZHU Z, et al. Spatio-temporal graph convolutional networks:a deep learning framework for traffic forecasting[EB/OL].(2017-09-04)[2025-03-01].https：//arxiv. org/abs/1709. 04875.
[31]VASWANI A, SHAZEER N, PARMAR N, et al. Attention is all you need[J]. Advances in neural information processing systems, 2017, 30:5998-6008.
[32]ADEPEJU M, ROSSER G, CHENG T. Novel evaluation metrics for sparse spatio-temporal point process hotspot predictions-a crime case study[J]. International journal of geographical information science, 2016, 30(11):2133-2154.
[33]中国裁判文书网[EB/OL].[2025-03-16]. https：//wenshu. court. gov. cn/.
[34]张全涛.大数据侦查运行机理研究[J].中国人民警察大学学报，2023,39(8):5-13.
[35]乔雅平.信息集成技术在城市应急联动系统建设中的应用[J].武警学院学报，2007,23(4):84-87.
基本信息:
中图分类号:D917
引用信息:
[1]陈晨,蔡欣怡.基于GIS技术的多特征犯罪时空序列预测模型构建[J].中国人民警察大学学报,2026,42(06):31-42.
2026-06-08
2026-06-08