基于主成分回归算法的城市客流聚集风险预测

doi:10.3969/j.issn.0255-8297.2019.04.010

应用科学学报 ›› 2019, Vol. 37 ›› Issue (4): 529-540.doi: 10.3969/j.issn.0255-8297.2019.04.010

基于主成分回归算法的城市客流聚集风险预测

王聚全^1,2, 王伟³, 马慧民⁴, 杨博¹, 杜渂^1,2

1. 迪爱斯信息技术股份有限公司, 上海 200032;
2. 电信科学技术第一研究所有限公司, 上海 200032;
3. 上海市公安局科技处, 上海 200042;
4. 上海市北高新股份有限公司, 上海 200436

收稿日期:2018-11-26 修回日期:2018-12-18 出版日期:2019-07-31 发布日期:2019-10-11
通信作者: 杜渂,教授级高工,研究方向:大数据、数据挖掘、物联网、软件体系架构,E-mail:duwen@dscomm.com.cn E-mail:duwen@dscomm.com.cn
基金资助:
工业和信息化部2018年大数据产业发展试点示范项目基金；上海市人工智能创新发展专项基金（No.2018-RGZN-01013）；上海市科技支撑计划项目基金（No.15dz1207400）；上海市信息化发展专项资金（No.201502001）；上海市科技创新行动计划项目基金（No.16511101000）资助

Urban Passenger Flow Aggregation Risk Forecasting Based on Principal Component Regression Algorithm

WANG Juquan^1,2, WANG Wei³, MA Huimin⁴, YANG Bo¹, DU Wen^1,2

1. DS Information Technology Co., Ltd., Shanghai 200032, China;
2. The First Research Institute of Telecommunications Technology Co., Ltd., Shanghai 200032, China;
3. Science and Technology Department of Shanghai Public Security Bureau, Shanghai 200042, China;
4. Shanghai Shibei Hi-tech Co., Ltd., Shanghai 200436, China

Received:2018-11-26 Revised:2018-12-18 Online:2019-07-31 Published:2019-10-11

摘要/Abstract

摘要： 为解决特大城市突发公共事件客流聚集预警准确度不高的问题，采用主成分回归算法对固定区域内运营商提供的手机用户数据和真实客流数据进行回归拟合，并根据多种统计检验方法对模型进行检验与评估.通过主成分分析，有效克服了运营商提供的手机用户数据的多重共线性问题，充分利用了手机用户数据的各个维度信息，并且降低了算法的复杂度，有效提高了城市客流聚集风险预测的精准度.

关键词: 主成分分析, 多重共线性, 统计检验, 客流聚集

Abstract: In order to solve the problem of the low accuracy of early warning of public events in mega-cities, this paper proposes a principal component regression algorithm to fit the mobile user data and real passenger flow data of fixed regions provided by operators, and uses a variety of statistical methods to test and evaluate the model. The principal component analysis can effectively overcome the multicollinearity problem of the mobile phone user data provided by the operators. By making full use of all-dimension information of the mobile phone user data, the complexity of the algorithm is reduced, and the accuracy of the prediction of the urban passenger flow aggregation risk is effectively improved.

Key words: principal component analysis, multicollinearity, statistical test, passenger flow aggregation

中图分类号:

TP311

王聚全, 王伟, 马慧民, 杨博, 杜渂. 基于主成分回归算法的城市客流聚集风险预测[J]. 应用科学学报, 2019, 37(4): 529-540.

WANG Juquan, WANG Wei, MA Huimin, YANG Bo, DU Wen. Urban Passenger Flow Aggregation Risk Forecasting Based on Principal Component Regression Algorithm[J]. Journal of Applied Sciences, 2019, 37(4): 529-540.

参考文献

[1] 杜渂,王聚全. 基于多源数据融合算法的城市客流聚集风险监测系统研究[J]. 网络安全技术与应用,2016(5):37-38. Du W, Wang J Q. Research on urban passenger flow aggregation risk monitoring system based on multi-source data fusion algorithm[J]. Network Security Technology and Application, 2016(5):37-38. (in Chinese)
[2] 孙超,吴宗之. 公共场所踩踏事故分析[J]. 安全,2007(1):18-23. Sun C, Wu Z Z. Analysis of trampling accidents in public places[J]. Safety, 2007(1):18-23. (in Chinese)
[3] 任常兴,吴宗之,刘茂. 城市公共场所人群拥挤踩踏事故分析[J]. 中国安全科学学报,2005, 15(12):102-106. Ren C X, Wu Z Z, Liu M. Analysis of crowded and trampling accidents in urban public places[J]. Chinese Journal of Safety Science, 2005, 15(12):102-106. (in Chinese)
[4] 冉丽君,刘茂. 人群密度对人群拥挤事故的影响[J]. 安全与环境学报,2007, 7(4):135-138. Ran L J, Liu M. The impact of crowd density on crowded accidents[J]. Journal of Safety and Environment, 2007, 7(4):135-138. (in Chinese)
[5] 白锐,梁力达,田宏. 人群聚集场所拥挤踩踏事故原因分析与对策[J]. 工业安全与环保, 2009, 35(2):47-49. Bai R, Liang L D, Tian H. Cause analysis and countermeasure of crowded and stamped accidents in crowd-collecting places[J]. Industrial Safety and Environmental Protection, 2009, 35(2):47-49. (in Chinese)
[6] 牟敬锋,赵星,樊静洁. 基于ARIMA模型的深圳市空气质量指数时间序列预测研究[J]. 环境卫生学杂志,2017(2):102-107. Mou J F, Zhao X, Fan J J. Time series prediction of Shenzhen air quality index based on ARIMA model[J]. Journal of Environmental Hygiene, 2017(2):102-107. (in Chinese)
[7] Lou P, Wu X F, Zhang X, Xu J B, Wang K. The epidemic analysis of brucellosis in Xinjiang base on the multiple seasonal ARIMA model[J]. Journal of Xinjiang Medical University, 2017(1):86-90.
[8] Steiner J E, Menezes R B, Ricci T V, Oliveira A S. PCA tomography:how to extract information from datacubes[J]. Monthly Notices of the Royal Astronomical Society, 2018, 395(1):64-75.
[9] Akaike H. Fitting autoregressive models for prediction[J]. Annals of the Institute of Statistical Mathematics, 1969, 21(1):243-247.
[10] Pun V C, Manjourides J, Suh H. Association of ambient air pollution with depressive and anxiety symptoms in older adults:results from the NSHAP study[J]. Environmental Health Perspectives, 2017, 125(3):342-348.
[11] Gang C, Xiao D, Zhang J, He Q, Wu X C. A new approach for online identification of low frequency oscillation modes based on auto-regressive moving-average model[J]. Power System Technology, 2010, 34(11):48-54.
[12] Lee S, Fambro D. Application of subset autoregressive integrated moving average model for short-term freeway traffic volume forecasting[J]. Journal of the Transportation Research Board, 1999, 1678(1):179-188.
[13] 孙云峰. 基于蒙特卡罗模拟的空间自相关系数检验[D]. 南京:南京师范大学,2006.
[14] 孙红果,邓华. 样本自相关系数与偏自相关系数的研究[J]. 蚌埠学院学报,2016(1):35-39. Sun H G, Deng H. Study on sample autocorrelation coefficient and partial autocorrelation coefficient[J]. Journal of Bengbu University, 2016(1):35-39. (in Chinese)
[15] Akaike H. Akaike's information criterion[M]. Heidelberg:Springer, 2011:25.
[16] Miller F P, Vandome A F, Mcbrewster J. Bayesian information criterion[J]. Ricksawatzky Com., 2014.
[17] 郭媛媛. 基于核主成分回归的多重共线性消除问题研究[D]. 唐山市:河北联合大学,2014.
[18] 高惠旋. 处理多元线性回归中自变量共线性的几种方法[J]. 数理统计与管理,2000, 20(5):204-207. Gao H X. Several methods for dealing with covariance of independent variables in multiple linear regression[J]. Mathematical Statistics and Management, 2000, 20(5):204-207. (in Chinese)
[19] 何方国,齐欢. 基于主成分分析与神经网络的非线性评价模型[J]. 武汉理工大学学报,2007, 29(8):183-186. He F G, Qi H. Nonlinear evaluation model based on principal component analysis and neural network[J]. Journal of Wuhan University of Technology, 2007, 29(8):183-186. (in Chinese)
[20] 张文霖. 主成分分析在SPSS中的操作应用[J]. 市场研究,2005(12):31-34. Zhang W L. Application of principal component analysis in SPSS[J]. Market Research, 2005(12):31-34. (in Chinese)
[21] 邓华玲,傅丽芳,孟军. 概率论与数理统计课程的改革与实践[J]. 大学数学,2004, 20(1):34-37. Deng H L, Fu L F, Meng J. Reform and practice of probability theory and mathematical statistics course[J]. College Mathematics, 2004, 20(1):34-37. (in Chinese)
[22] Baker L D, Frank L L, Fosterschubert K, Green P S, Wilkinson C W, Mctiernan A, Cholerton B A, Plymate S R, Fishel M A, Watson G S, Duncan G E, Mehta P D, Craft S. Aerobic exercise improves cognition for older adults with glucose intolerance, a risk factor for Alzheimer's disease[J]. Journal of Alzheimers Disease, 2016, 22(2):569-579.
[23] Engle R F. Autoregressive conditional heteroscedasticity with estimates of the variance of United Kingdom inflation[J]. Econometrica, 1982, 50(4):987-1007.

基于主成分回归算法的城市客流聚集风险预测

Urban Passenger Flow Aggregation Risk Forecasting Based on Principal Component Regression Algorithm

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 8

编辑推荐

Metrics

本文评价

[1]	王金伟, 徐春晖. 基于QPCA的彩色图像复制粘贴检测[J]. 应用科学学报, 2019, 37(2): 291-300.
[2]	马丹, 程辉, 毛艳玲, 邢世和. 基于DMSP/OLS夜间灯光影像的省际经济发展水平评估模型研究[J]. 应用科学学报, 2017, 35(5): 647-657.
[3]	王向阳1,2, 万旺根1,2. 采用多任务稳健主成分分析的运动目标分割[J]. 应用科学学报, 2014, 32(5): 473-480.
[4]	孟一飞1，袁雪1，魏学业1，秦飞舟2，覃庆努1. 基于样本扩展和线性子空间特征提取的单样本人脸识别[J]. 应用科学学报, 2013, 31(5): 488-494.
[5]	陈波，蔡锦霖，陈文胜，刘振河，李妍. 一种基于2DPCA训练的形状先验提取方法[J]. 应用科学学报, 2013, 31(1): 84-90.
[6]	马飞虎1;2，孙翠羽1;3，康永红1，刘智敏3. 多特征主成分分析与声图相结合的海底底质分类[J]. 应用科学学报, 2010, 28(4): 374-380.
[7]	胡晓俞王新姚菁. 利用二元结构特征的人脸识别[J]. 应用科学学报, 2010, 28(3): 271-276.
[8]	王玥，王树根. 高分辨率遥感影像阴影检测与补偿的主成分分析方法[J]. 应用科学学报, 2010, 28(2): 136-141.