[Purpose/Significance] The number of serious violent crimes is an important indicator to evaluate the overall level of public security, and the use of information science methods to accurately predict the number of cases based on social macro data is of great significance for scientifically allocating security and prevention resources, accurately preventing and combating crimes and improving social governance capabilities. [Method/Process] Based on the perspective of social macro data, the basic principles of criminology were used to deeply analyze the complex influencing factors of serious violent crimes, and the research hypothesis of the correlation factors and prediction methods of the number of serious violent crimes was proposed, and the corresponding social macro data were selected as variables, then the autoregression lag model was established by combining the historical occurrence data of eight types of serious violent crimes in the past 20 years, and the incidence pattern of serious violent crimes was predicted. [Result/Conclusion] The study finds that macro social indicators such as GDP, marriage rate, population structure, security input, and unemployment rate have a significant correlation with the number of serious violent crimes, and the accuracy of prediction is greatly improved compared with the single use of historical data, which provides strong technical support and methodological guidance for China’s scientific formulation of relevant social governance prevention strategies.
[1] 中华人民共和国国民经济和社会发展第十四个五年规划和2035年远景目标纲要[N].人民日报, 2021-03-13(1).
[2] 公安部.二十大记者招待会:公安机关扎实推进平安中国、法治中国建设[EB/OL].[2023-03-04]. https://www.mps.gov.cn/n2253534/n2253535/c8736964/content.html.
[3] 靳高风,张雍锭,郭兆轩. 2021-2022年中国犯罪形势分析与预测[J].中国人民公安大学学报(社会科学版), 2022, 38(2):1-12.
[4] 曹子丹.《中国犯罪原因研究综述》[M].北京:中国政法大学出版社, 1993.
[5] 吴之辉,丁红军,尚欣.个人极端暴力犯罪特点、原因及防范对策[J].中国公共安全(学术版), 2019(3):33-36.
[6] 胡铭,严敏姬.大数据视野下犯罪预测的机遇、风险与规制——以英美德"预测警务"为例[J].西南民族大学学报(人文社会科学版), 2021, 42(12):84-91.
[7] 屈茂辉,郝士铭.基于ARMA模型的我国财产类犯罪人数预测研究[J].中国刑事法杂志, 2013(4):100-106.
[8] 王璟瑶,秦静,刘军.犯罪案件移动平均预测模型改进研究[J].中国人民公安大学学报(自然科学版), 2015, 21(1):65-67.
[9] 颜靖华,侯苗苗.基于LSTM网络的盗窃犯罪时间序列预测研究[J].数据分析与知识发现, 2020, 4(11):84-91.
[10] 钟飚,袁梦佳.犯罪时间序列预测分析方法研究——以CrimeStat软件为例[J].中国人民公安大学学报(自然科学版), 2019, 25(2):58-63.
[11] HASSANI H, XU H, SILVA E S, et al. A review of data mining applications in crime[J]. Statistical analysis and data mining, 2016, 9(3):139-154.
[12] 朱小波,次晋芳.基于改进PSO-BP神经网络算法在一般盗窃犯罪预测中的应用[J].计算机应用与软件, 2020, 37(1):37-42, 75.
[13] 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.
[14] 陈晨,李勇男,王铭戬.基于Lightgbm算法的恐怖袭击预警模型构建研究[J].情报杂志, 2022, 41(6):21-28, 98.
[15] 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.
[16] SALTOS G, COCEA M. An exploration of crime prediction using data mining on open data[J]. International journal of information technology and decision making, 2017, 16(5):1155-1181.
[17] 邱明月,王新猛,唐松泽.基于人工神经网络模型的抢劫犯罪微观研究[J].信息技术与信息化, 2018(10):140-143.
[18] KHAIRUDDIN A R, ALWEE R, HARON H. A comparative analysis of artificial intelligence techniques in forecasting violent crime rate[J]. IOP conference series materials science and engineering, 2020, 864:012056.
[19] 张强,何乐平.基于支持向量机的治安高危人员风险预警方法[J].管理评论, 2016, 28(11):245-251.
[20] 陈鹏,胡啸峰,陈建国.基于模糊信息粒化的支持向量机在犯罪时序预测中的应用[J].科学技术与工程, 2015, 15(35):54-57, 63.
[21] 翟圣昌,韩晓红,王莉,等.基于BP神经网络非线性组合的SARIMA-GRU犯罪预测模型[J/OL].太原理工大学学报:1-13[2022-10-28]. http://kns.cnki.net/kcms/detail/14.1220.n.20220705.1240.002.html.
[22] 朱小波,栗赫遥.基于PCA-XGBoost联合算法的盗窃犯罪预测模型[J].计算机应用与软件, 2022, 39(5):98-103.
[23] 翟一鸣,丁宁,李成龙.基于节假日因素的多尺度犯罪时序预测方法研究[J].中国人民公安大学学报(自然科学版), 2020, 26(3):78-84.
[24] 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.
[25] 唐德权,史伟奇,张波云.基于多模态信息特征融合的犯罪预测算法研究[J].计算机应用与软件, 2018, 35(7):221-225, 262.
[26] 夏勇.犯罪本质特征新论[J].法学研究, 2001(6):3-21.
[27] ROSENZWEIG S. Types of reaction to frustration[J]. The journal of abnormal and social psychology, 1934, 29(3):298-300.
[28] DOLLARD J, DOOB L W, MILLER N E, et al. Frustration and aggression[J]. American journal of sociology, 1939, 92(7):1654-1667.
[29] MERTON R K. Social theory and social structure[M]. New York:Free Press, 1968.
[30] GRANT D. The patterns and prevalence of mass murder in twentieth-century America[J]. Justice quarterly, 2004, 21(4):729-761.
[31] 于小川,刘江斌.个人极端暴力犯罪的形成机理与逻辑推演[J].湖南警察学院学报, 2021, 33(4):23-31.
[32] 靳高风,李易尚,朱双洋.当前我国校园个人极端暴力犯罪研究——基于2001-2016年间31个典型案件的调查[J].山东警察学院学报, 2017, 29(3):9.
[33] HILL H M, LEVERMORE M, TWAITE J, et al. Exposure to community violence and social support as predictors of anxiety and social and emotional behavior among African American children[J]. Journal of child&family studies, 1996, 5(4):399-414.
[34] 刘跃威.我国社会转型时期个人极端暴力犯罪探究[J].山西省政法管理干部学院学报, 2020, 33(2):83-86.
[35] 冯卫国.个人极端暴力犯罪的概念厘定与类型界分[J].犯罪研究, 2021(4):2-13.
[36] KWON R, CABRERA J F. Income inequality and mass shootings in the United States[J]. BMC public health, 2019.9(19):1-8.
[37] CABRERA J F, KWON R. Income inequality, household income, and mass shooting in the United States[J]. Frontiers in public health, 2018, 10(6):1-4.
[38] 李培林.另一只看不见的手:社会结构转型[J].中国社会科学, 1992(5):3-17.
[39] 张威.改革开放以来我国犯罪问题的宏观分析[M].北京:中国政法大学出版社, 2018.
