|本期目录/Table of Contents|

[1]邱凌峰,胡啸峰,周睿,等.基于机器学习的典型社会安全事件发生规律研究及对雄安新区的启示[J].中国安全生产科学技术,2018,14(10):11-17.[doi:10.11731/j.issn.1673-193x.2018.10.002]
　QIU Lingfeng,HU Xiaofeng,ZHOU Rui,et al.Research on the occurrence regularity of typical social security incidents based on machine learning and implications for Xiongan New Area[J].JOURNAL OF SAFETY SCIENCE AND TECHNOLOGY,2018,14(10):11-17.[doi:10.11731/j.issn.1673-193x.2018.10.002]
点击复制

基于机器学习的典型社会安全事件发生规律研究及对雄安新区的启示()
分享到：

《中国安全生产科学技术》[ISSN:1673-193X/CN:11-5335/TB]

卷:
14
期数:
2018年10期
页码:
11-17
栏目:
特邀专栏
出版日期:
2018-10-31

文章信息/Info

Title:
Research on the occurrence regularity of typical social security incidents based on machine learning and implications for Xiongan New Area
文章编号:
1673-193X(2018)-10-0011-07
作者:
邱凌峰1; 2; 胡啸峰1; 2; 周睿3; 4;  顾海硕1; 2; 唐正1; 2; 郑超慧1; 2; 张学军1; 2
(1.中国人民公安大学 信息技术与网络安全学院，北京 102623；2.安全防范技术与风险评估公安部重点实验室，北京 102623；3.清华大学 工程物理系，北京 100084；4.清华大学 公共安全研究院，北京 100084）
Author(s):
QIU Lingfeng1; 2;  HU Xiaofeng1; 2;  ZHOU Rui3; 4; GU Haishuo1; 2; TANG Zheng1; 2;  ZHENG Chaohui1; 2 ; ZHANG Xuejun1; 2
(1. School of Information Technology and Cyber Security, People’s Public University of China, Beijing 102623, China;2. Key Laboratory ofSecurity Technology & Risk Assessment, Ministry of public security, Beijing 102623, China;3. Department of Engineering Physics，Tsinghua University，Beijing 100084，China;4. Institute of Public Safety Research，Tsinghua University，Beijing 100084，China)
关键词:
雄安新区; 社会安全; 机器学习; 分类预测; 社会治安防控
Keywords:
Xiongan New Area;  social security;  machine learning;  classification and prediction;  social security prevention and control
分类号:
X913；TP181
DOI:
10.11731/j.issn.1673-193x.2018.10.002
文献标志码:
A
摘要:
针对雄安新区建设和发展过程中对社会安全事件的防控需求，以盗窃作为典型社会安全事件，提出基于机器学习模型的社会安全事件分析预测方法，并以A市2012—2016年的实际盗窃犯罪数据为基础，提取发案时间、发案地点、实施手段和损失金额作为分类特征，通过比较多种机器学习算法，研究盗窃前科人员的预测方法，并根据预测结果挖掘盗窃前科人员的作案规律。研究结果表明：随机森林算法表现最优,查准率、查全率和F1均达到了0.85以上；对于盗窃这一典型社会安全事件，其前科人员倾向于选择下午时段和人流量大的地区实施，盗窃金额明显高于初犯和惯犯。最后，基于前述研究，提出构建数据驱动的社会安全事件预测预警和综合研判系统，并针对该系统的前期建设和后期使用，给出“制定统一的数据格式”、“实现数据实时接入”的建议。相关研究成果可为雄安新区社会安全事件预测预警以及治安防控工作的开展提供参考和借鉴。
Abstract:
Aiming at the need of prevention and control of social security incidents in the construction and development of Xiongan New Area, taking theft as a typical social security incidents, methods of social security incidents analysis and prediction based on machine learning model are proposed. The methods of prediction of larceny exconvict were studied using the larceny data in city A from 2012 to 2016, based on six kinds of Machine Learning Models, extracting the duration, location, means and loss of the victim as the classification features. Moreover, the laws of crime commitment by larceny exconvict were examined. The results show that Random Forest has the best performance on prediction of larceny exconvict, which has the highest precision, recall and F1-score as 0.85. For theft, a typical social security incident, larceny exconvict usually commit crimes by selecting the location where the flow of people is large in the afternoon. The loss of the victim resulted from larceny exconvict are evidently more than that from first offenders and recidivists. Finally, based on the above machine learning algorithm ,a prediction and early warning and comprehensive evaluation system for social security incidents based on data guiding is proposed. In view of the early construction and later use of the system, two suggestions are given, which are “developing a unified data format” and “realizing data realtime access”. The relevant research results of this paper can provide a reference for the prediction and early warning of social security incidents and social security prevention and control in Xiongan New Area.

参考文献/References:

［1］周睿, 陈鹏, 胡啸峰,等. 雄安新区城市化进程中人口风险驱动的社会治理策略初探［J］. 中国安全生产科学技术, 2018,14(8)：5-11.ZHOU Rui, CHEN Peng,HU Xiaofeng, et al.Primary exploration of social governance strategies driven by population risk during urbanization process of Xiongan New Area［J］. Journal of Safety Science and Technology,2018,14(8):5-11.
［2］靳高风,朱双洋,林晞楠.中国犯罪形势分析与预测(2017-2018)［J］.中国人民公安大学学报(社会科学版),2018,34(2):29-38.JIN Gaofeng,ZHU Shuangyang,LIN Xinan.Analysis and prediction of criminal situation in China (2017-2018) ［J］. Journal of Chinese People’s Public Security University(Social Sciences Edition), 2018,34(2): 29-38.
［3］吕雪梅. 美国犯罪情报预测分析技术的特点——基于兰德报告《预测警务》的视角［J］. 情报杂志, 2016, 35(7):7-12.LYU Xuemei.The Characteristics of the Criminal Intelligence Predictive Analytical Technology in the United States: an Analysis from Rand’s Report［J］. Journal of Intelligence,2016,35(7): 7-12.
［4］李国军. 论大数据驱动下的预测警务创新［J］. 中国人民公安大学学报(社会科学版), 2015, 31(6):3-8.LI Guojun.On the prediction of police innovation driven by big data［J］. Journal of Chinese People’s Public Security University(Social Sciences Edition), 2015,31(6): 3-8.
［5］TOLLENAAR N,HEIJDEN P G M V D. Which method predicts recidivism best?: a comparison of statistical, machine learning and data mining predictive models［J］. Journal of the Royal Statistical Society, 2013, 176(2):565-584.
［6］PFLUEER M O,FRANKE I,GRAF M, et al. Predicting general criminal recidivism in mentally disordered offenders using a random forest approach［J］. Bmc Psychiatry, 2015, 15(1):1-10.
［7］VURAL M S, Gk M. Criminal prediction using Naive Bayes theory［J］. Neural Computing & Applications, 2016, 28(9):1-12.
［8］李雨聪,刘硕,王方明.基于环境犯罪学的犯罪预测模型的建立［J］.情报杂志,2018,37(2):45-49,56.LI Yucong,LIU Shuo,WANG Fangming.The Prediction model for crime based on the environmental criminology［J］. Journal of Intelligence,2018,37(2): 45-49,56.
［9］李荣岗, 孙春华, 姬建睿. 基于支持向量机的嫌疑人特征预测［J］. 计算机工程, 2017, 43(11):198-203.LI Ronggang,SUN Chunhua,JI Jianrui.Suspect characteristics prediction based on support vector machine［J］. Computer Engineering. 2017, 43(11): 198-203.
［10］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.LI Weihong,WEN Lei,CHEN Yebin.Property crime forecast based on improved GA-BP neural network model［J］. Geomatics and Information Science of Wuhan University,2017,42(8): 1110-1116.
［11］孙菲菲,曹卓,肖晓雷.基于随机森林的分类器在犯罪预测中的应用研究［J］.情报杂志,2014,33(10):148-152.SUN Feifei,CAO Zhuo,XIAO Xiaolei.Application of an improved random forest based classifier in crime prediction domain［J］. Journal of Intelligence, 2014,33(10): 148-152.
［12］Nitesh V. Chawla, Kevin W. Bowyer, Lawrence O. Hall, et al. SMOTE: synthetic minority over-sampling technique［J］. Journal of Artificial Intelligence Research, 2011, 16(1):321-357.
［13］韩慧, 王路, 温明,等. 不均衡数据集学习中基于初分类的过抽样算法［J］. 计算机应用, 2006, 26(8):1894-1897.HAN Hui,WANG Lu,WEN Ming, et al.Over-sampling algorithm based on classification of imbalance data sets learning［J］. Journal of Computer Applications,2006,26(8): 1894-1897.
［14］KEVIN S. WOODS, CHRISTOPHER C. DOSS, KEVIN W. BOWYER, et al. Comparative evaluation of pattern recognition techniques for detection of microcalcifications in mammography［J］. International Journal of Pattern Recognition and Artificial Intelligence, 1993, 7(6):1417-1436.
［15］BLAGUS R,LUSA L. SMOTE for high-dimensional class-imbalanced data.［J］. BMC Bioinformatics,14,1(2013-03-22), 2013, 14(1):1-16.
［16］周志华. 机器学习［M］.北京: 清华大学出版社, 2016.
［17］钟雅, 郭渊博. 基于机器学习的日志解析系统设计与实现［J］. 计算机应用, 2018, 38(2):352-356.ZHONG Ya,GUO Yuanbo.Design and implementation of log parsing system based on machine learning［J］. Journal of Computer Applications,2018,38(2): 352-356.
［18］刘庆生, 刘高焕, 蔺启忠,等. 基于逻辑斯蒂模型的遥感图像分类［J］. 国土资源遥感, 2001, 13(1):36-41.LIU Qingsheng,LIU Gaohuan,LIN Qizhong, et al. Remote sensing image classification based on logistic model［J］. Remote Sensing for Land & Resources,2001,13(1): 36-41.
［19］周于皓, 张红玲, 李芳菲,等. 局部关注支持向量机算法［J］. 计算机应用, 2018, 38(4):945-948.ZHOU Yuhao,ZHANG Hongling,LI Fangfei, et al.Local focus support vector machine algorithm［J］. Journal of Computer Applications,2018,38(4): 945-948.
［20］FRIEDL M A,BRODLEY C E. Decision tree classification of land cover from remotely sensed data［J］. Remote Sensing of Environment, 1997, 61(3):399-409.
［21］BREIMAN L. Random Forests［J］. Machine Learning, 2001, 45(1):5-32.
［22］DENOEUX T. A k-Nearest neighbor classification rule based on dempster-shafer theory［J］. Systems Man & Cybernetics IEEE Transactions on, 1995, 25(5):804-813.
［23］WANG Q,GARRITY M,TIEDJE J M, et al. Naive Bayesian classifier for rapid assignment of rRNA sequences into the new bacterial taxonomy［J］. Applied & Environmental Microbiology,2007, 73(16): 5264-5267.

相似文献/References:

[1]耿红杰,周海彬,穆帅伟.海外石油工程项目社会安全应急资源 管理实践与思考[J].中国安全生产科学技术,2012,8(7):133.
　GENG Hong jie,ZHOU Hai bin,MU Shuai wei.Practice and thinking of security emergency resources management  in overseas petroleum engineering projects[J].JOURNAL OF SAFETY SCIENCE AND TECHNOLOGY,2012,8(10):133.
[2]黄弘,李瑞奇,范维澄,等.安全韧性城市特征分析及对雄安新区安全发展的启示[J].中国安全生产科学技术,2018,14(7):5.[doi:10.11731/j.issn.1673-193x.2018.07.001]
　HUANG Hong,LI Ruiqi,FAN Weicheng,et al.Analysis on characteristics of safety resilient city and enlightenments for safe development of Xiongan New Area[J].JOURNAL OF SAFETY SCIENCE AND TECHNOLOGY,2018,14(10):5.[doi:10.11731/j.issn.1673-193x.2018.07.001]
[3]赵鹏霞,朱伟,王亚飞.韧性社区评估框架与应急体制机制设计及在雄安新区的构建路径探讨[J].中国安全生产科学技术,2018,14(7):12.[doi:10.11731/j.issn.1673-193x.2018.07.002]
　ZHAO Pengxia,ZHU Wei,WANG Yafei.Assessment framework and emergency system and mechanism design of resilient community and discussion on its construction path in Xiongan New Area[J].JOURNAL OF SAFETY SCIENCE AND TECHNOLOGY,2018,14(10):12.[doi:10.11731/j.issn.1673-193x.2018.07.002]
[4]周睿,陈鹏,胡啸峰,等.雄安新区城市化进程中人口风险驱动的社会治理策略初探[J].中国安全生产科学技术,2018,14(8):5.[doi:10.11731/j.issn.1673-193x.2018.08.001]
　ZHOU Rui,CHEN Peng,HU Xiaofeng,et al.Primary exploration of social governance strategies driven by population risk during urbanization process of Xiongan New Area[J].JOURNAL OF SAFETY SCIENCE AND TECHNOLOGY,2018,14(10):5.[doi:10.11731/j.issn.1673-193x.2018.08.001]
[5]钟茂华,孟洋洋.安全生产韧性管理对雄安新区发展的借鉴[J].中国安全生产科学技术,2018,14(8):12.[doi:10.11731/j.issn.1673-193x.2018.08.002]
　ZHONG Maohua,MENG Yangyang.Reference of Work Safety Resilient Management to the Development of Xiongan New Area[J].JOURNAL OF SAFETY SCIENCE AND TECHNOLOGY,2018,14(10):12.[doi:10.11731/j.issn.1673-193x.2018.08.002]
[6]陈涛,王玉井.安全韧性雄安新区中的卫生应急风险与对策研究[J].中国安全生产科学技术,2018,14(8):18.[doi:10.11731/j.issn.1673-193x.2018.08.003]
　CHEN Tao,WANG Yujing.Study on risk and countermeasures of public health emergency in safety resilient Xiongan New Area[J].JOURNAL OF SAFETY SCIENCE AND TECHNOLOGY,2018,14(10):18.[doi:10.11731/j.issn.1673-193x.2018.08.003]
[7]吴平,林浩曦,田璐.基于生态系统服务供需的雄安新区生态安全格局构建[J].中国安全生产科学技术,2018,14(9):5.[doi:10.11731/j.issn.1673-193x.2018.09.001]
　WU Ping,LIN Haoxi,TIAN Lu.Construction of ecological security pattern based on: A case study in Xiongan New Area, Hebei Province, China[J].JOURNAL OF SAFETY SCIENCE AND TECHNOLOGY,2018,14(10):5.[doi:10.11731/j.issn.1673-193x.2018.09.001]
[8]陈志芬,李俊伟,卢方欣,等.城市消防站选址布局优化及对雄安新区的启示[J].中国安全生产科学技术,2018,14(9):12.[doi:10.11731/j.issn.1673-193x.2018.09.002]
　CHEN Zhifen,LI Junwei,LU Fangxin,et al.Optimizing location of fire stations and it’s enlightenments for Xiongan New Area[J].JOURNAL OF SAFETY SCIENCE AND TECHNOLOGY,2018,14(10):12.[doi:10.11731/j.issn.1673-193x.2018.09.002]
[9]李维明,何凡,谷树忠.雄安新区水安全治理的对策建议研究[J].中国安全生产科学技术,2018,14(10):5.[doi:10.11731/j.issn.1673-193x.2018.10.001]
　LI Weiming,HE Fan,GU Shuzhong.Research on key areas and measures of water safetygovernance in Xiongan New Area[J].JOURNAL OF SAFETY SCIENCE AND TECHNOLOGY,2018,14(10):5.[doi:10.11731/j.issn.1673-193x.2018.10.001]

备注/Memo

备注/Memo:
收稿日期： 2018-10-11
基金项目： 国家自然科学基金项目（71741027,71704183）；国家重点研发计划课题(2018YFC0809702)；公安部科技强警基础工作专项项目(2018GABJC01)
作者简介： 邱凌峰，硕士研究生，主要研究方向为机器学习。
通信作者： 胡啸峰，博士，讲师，主要研究方向为警务信息技术。
更新日期/Last Update: