Please wait a minute...
浙江大学学报(工学版)
浙江大学学报(工学版)  2019, Vol. 53 Issue (9): 1759-1767    DOI: 10.3785/j.issn.1008-973X.2019.09.015
计算机科学与人工智能     
多源数据跨国人口迁移预测
汪子龙(),王柱*(),於志文,郭斌,周兴社
西北工业大学 计算机学院，陕西 西安 710129
Transnational population migration forecast with multi-source data
Zi-long WANG(),Zhu WANG*(),Zhi-wen YU,Bin GUO,Xing-she ZHOU
College of Computer Science, Northwestern Polytechnical University, Xi'an 710129, China
 全文: PDF(925 KB)   HTML
摘要：

针对跨国人口迁移预测所面临的数据代表性问题，利用多源数据分别构建3个预测模型：线性拟合模型、乘法分量模型和带有时间序列预测（WTSP）的线性拟合模型. 线性拟合模型用于刻画1年内的移民规律；乘法分量模型利用时间序列预测算法对未来迁移模式进行预测；WTSP线性拟合模型利用迁移模式的变化预测跨国人口迁移数量的未来趋势. 对比3个模型的预测结果可知，WTSP线性拟合模型可以有效预测未来的移民规律，相比经典线性拟合模型，WTSP线性拟合模型能体现迁移模式随时间变化的规律，预测准确率可至少提升3%；相比乘法分量模型，WTSP线性拟合模型能呈现更完整的迁移模式，有更强的可解释性.
关键词： 人口迁移预测;  回归分析;  时间序列预测;  线性拟合;  乘法分量模型;  WTSP线性拟合模型    
Abstract:

Three prediction models were constructed by using multi-source data, including linear fitting model, multiplicative component model and WTSP (with time series prediction) linear fitting model, aiming at the problem of data representativeness in the prediction of transnational migration. A linear fitting model was constructed to describe the migration rule within one year; a multiplier component model was introduced to predict the future migration pattern by using time series prediction algorithm; and a WTSP linear fitting model was proposed to predict the future trend of transnational migration by using the change of the migration pattern. Compared with the results of the three models, WTSP linear fitting model can effectively predict future migration patterns. Compared with the classical linear fitting model, the WTSP linear fitting model can reflect the law of migration pattern changing with time, and the prediction accuracy can be improved by at least 3%. Compared with the multiplier component model, the WTSP linear fitting model can present a more complete migration model and has stronger interpretability.
Key words: population migration prediction    regression analysis    time series prediction    linear fitting    multiplicative component model    WTSP linear fitting model
收稿日期: 2018-12-17 出版日期: 2019-09-12
CLC:  TP 399  
通讯作者: 王柱     E-mail: 1977431951@qq.com;wangzhu@nwpu.edu.cn
作者简介: 汪子龙（1996—），男，硕士生，从事普适计算和数据挖掘研究. orcid.org/0000-0003-4471-872X. E-mail: 1977431951@qq.com
服务　　
把本文推荐给朋友
加入引用管理器
E-mail Alert
作者相关文章　　
汪子龙
王柱
於志文
郭斌
周兴社

引用本文:

汪子龙,王柱,於志文,郭斌,周兴社. 多源数据跨国人口迁移预测[J]. 浙江大学学报(工学版), 2019, 53(9): 1759-1767.

Zi-long WANG,Zhu WANG,Zhi-wen YU,Bin GUO,Xing-she ZHOU. Transnational population migration forecast with multi-source data. Journal of ZheJiang University (Engineering Science), 2019, 53(9): 1759-1767.

链接本文:

http://www.zjujournals.com/eng/CN/10.3785/j.issn.1008-973X.2019.09.015    或    http://www.zjujournals.com/eng/CN/Y2019/V53/I9/1759

c d acd /人
英国 德国 23 574
英国 意大利 1 548
德国 英国 18 659
德国 意大利 1 877
意大利 英国 10 733
意大利 德国 20 083
表 1  部分国家间的移民数量
c d ucd
英国 德国 1.643 7
英国 意大利 1.375 4
德国 英国 2.364 3
德国 意大利 2.040 2
意大利 英国 0.906 3
意大利 德国 1.141 5
表 2  基于乘法分量模型的国家间吸引系数
图 1  长短期记忆(LSTM)单元结构图
特征 p值 特征 p值
邮件收取 0 旅游 0
邮件发出 0 劳动力 0.038 4
电话打入 0.531 8 经济 0
电话打出 0 人口 0.000 8
网络活动 0.345 5 语种 0.255 9
教育 0.016 5 ? ?
表 3  线性拟合模型特征选择中不同特征的p值结果
图 2  2013年欧盟12国与意大利间移民人数的线性拟合模型的预测值与真实值对比
年份 c Fc Tc ucI uIc
2008 英国 0.044 0.143 0.234 0.680
德国 0.062 0.272 0.236 1.385
荷兰 0.029 0.045 0.112 1.091
2009 英国 0.042 0.164 0.338 1.001
德国 0.063 0.314 0.291 1.257
荷兰 0.034 0.040 0.114 0.955
2010 英国 0.041 0.152 0.318 1.172
德国 0.058 0.322 0.301 1.158
荷兰 0.035 0.046 0.116 0.840
2011 英国 0.038 0.170 0.361 1.122
德国 0.059 0.309 0.301 1.163
荷兰 0.035 0.048 0.108 0.806
2012 英国 0.041 0.167 0.359 1.089
德国 0.053 0.317 0.336 1.075
荷兰 0.036 0.051 0.108 0.847
表 4  2008—2012年基于乘法分量模型的部分国家间吸引系数
图 3  2013年欧盟12国移入意大利人数的乘法分量模型预测值与真实值对比
图 4  2013年意大利移入欧盟12国人数的乘法分量模型预测值与真实值对比
图 5  2013年欧盟12国与意大利间移民人数WTSP线性拟合模型预测效果与线性拟合模型预测效果对比
图 6  2012年欧盟12国与意大利间移民人数的线性拟合模型预测值与真实值对比
1 BILSBORROW R E, HUGO G, OBERAI A S International migration statistics: guidelines for improving data collection systems[J]. International Labour Office, 1997, 33 (1): 204
2 COLEMAN D The twilight of the census[J]. Population and Development Review, 2013, 38 (Suppl.1): 334- 351
3 JAMES R, ARKADIUSZ W, JONATHAN J, et al Integrated modeling of European migration[J]. Journal of the American Statistical Association, 2013, 108 (503): 801- 819
doi: 10.1080/01621459.2013.789435
4 KUPISZEWSKA D, WI?NIOWSKI A Availability of statistical data on migration and migrant population and potential supplementary sources for data estimation[J]. Jornal Brasileiro de Patologia e Medicina Laboratorial, 2003, 43 (43): 235- 240
5 GROENEWOLD W G F, BILSBORROW R, BONIFAZI C, et al Design of samples for international migration surveys: methodological considerations and lessons learned from a multi-country study in Africa and Europe[J]. Imiscoe Research, 2008, 293- 312
6 ABEL G J Estimating global migration flow tables using place of birth data[J]. Demographic Research, 2013, 28 (2): 505- 546
7 ABEL G J Estimates of global bilateral migration flows by gender between 1960 and 2015[J]. International Migration Review, 2017, (11), 52 (3): 809- 852
8 ABEL G J, SANDER N Quantifying global international migration flows[J]. Science, 2014, 343 (6178): 1520- 1522
doi: 10.1126/science.1248676
9 HAWELKA B, SITKO I, BEINAT E, et al Geo-located Twitter as proxy for global mobility patterns[J]. Cartography and Geographic Information Science, 2014, 41 (3): 260- 271
doi: 10.1080/15230406.2014.890072
10 LENORMAND M, TUGORES A, COLET P, et al Tweets on the road[J]. PLoS One, 2014, 9 (8): e105407
doi: 10.1371/journal.pone.0105407
11 WANG W, DAVID R, SHARAD G, et al Forecasting elections with non-representative polls[J]. International Journal of Forecasting, 2015, 31 (3): 980- 991
doi: 10.1016/j.ijforecast.2014.06.001
12 RAYMER J, ABEL G, SMITH P W F Combining census and registration data to estimate detailed elderly migration flows in England and Wales[J]. Journal of the Royal Statistical Society: Series A (Statistics in Society), 2007, 170 (4): 891- 908
doi: 10.1111/rssa.2007.170.issue-4
13 DORIGO G, TOBLER W Push-pull migration laws[J]. Annals of the Association of American Geographers, 1983, 73 (1): 1- 17
doi: 10.1111/j.1467-8306.1983.tb01392.x
14 HU X, MANAGEMENT S O Analysis on the motivation and obstruction of in-situ urbanization in China based on the push and pull theory[J]. Journal of Hebei Normal University of Science and Technology, 2017, 16 (4): 38- 45
15 MARKOVSKY, IV AN, VAN H, et al Overview of total least-squares methods[J]. Signal Processing, 2013, 87 (10): 2283- 2302
16 ESCANCIANO J C Goodness-of-fit tests for linear and nonlinear time series models[J]. Publications of the American Statistical Association, 2006, 101 (474): 531- 541
doi: 10.1198/016214505000001050
17 ALI B N, DANNY H, LUIZ F C Towards an early soft-ware estimation using log-linear regression and a multilayer perceptron model[J]. The Journal of Systems and Software, 2013, 86 (1): 144- 160
doi: 10.1016/j.jss.2012.07.050
18 QIAO C, CHEN H B, JING W F, et al Towards establishing a meaningful and practical dynamics results for the unified RNN model[J]. Neurocomputing, 2015, 157: 315- 322
doi: 10.1016/j.neucom.2014.12.007
19 廖大强, 印鉴 基于多分支RNN快速学习算法的混沌时间序列预测[J]. 计算机应用研究, 2015, 32 (2): 403- 408
LIAO Da-qiang, YIN Jian Chaotic time series of fast learning algorithm of multi branch prediction based on RNN[J]. Application Research of Computers, 2015, 32 (2): 403- 408
doi: 10.3969/j.issn.1001-3695.2015.02.019
20 LI Y F, CAO H Prediction for tourism flow based on LSTM neural network[J]. Procedia Computer Science, 2018, 129: 227- 283
21 张亮, 黄曙光, 石昭祥, 等 基于LSTM型RNN的CAPTCHA识别方法[J]. 模式识别与人工智能, 2011, 24 (1): 40- 47
ZHANG Liang, HUANG Shu-guang, SHAO Zhao-xiang, et al CAPTCHA recognition method based on RNN of LSTM[J]. Pattern Recognition and Artificial Intelligence, 2011, 24 (1): 40- 47
doi: 10.3969/j.issn.1003-6059.2011.01.005
[1] 王晨霖,杨洁,居文军,顾复,陈芨熙,纪杨建. 基于智能家电的短期电力负荷预测与削峰填谷优化[J]. 浙江大学学报(工学版), 2020, 54(7): 1418-1424.
[2] 毛贺,陈章位,黄靖,胡科伟. PCR仪温度场热模型的研究与验证[J]. J4, 2013, 47(9): 1619-1624.