王森 肖渝 黄群英 张纯

Wang Sen, Xiao Yu, Huang Qunying, Zhang Chun

关键词：社交媒体；大数据；城市安全；减灾；数据挖掘

Keywords：Social Media; Big Data; Urban Security; Mitigation; Data Mining

摘要：

在城市灾害频发的背景下，社交媒体大数据在灾害分析中所能够发挥的作用得到了越来越多的关注。对于社交大数据的挖掘和使用，主要体现在诸如灾情感知、信息编码、事件跟踪、灾难救援以及损失评估等领域。本文以2012 年在美国多地特别是纽约市造成了严重影响的桑迪飓风为例，基于社交媒体网站推特（Twitter）以及相关数据库的信息，通过信息编码、分类以及空间网络的对接等方式，研究发现灾前准备、灾害发生、灾害响应和灾后应对等主题随时间、空间发展的趋势等特征。本文通过构建回归模型描述并讨论了与灾情相关的解释性变量同推文数量间的关系。与此同时，本文参照MMAM 理论①讨论了推文灾情与真实情况的误差产生原因。研究结果表明，推特信息的数量与人口规模和著名的地标性区域显著相关，个人属性如教育程度、年龄、性别等也对推特信息数量产生影响。本文希望通过对信息化背景下社交媒体大数据信息的挖掘和分析，从社交媒体信息发布特征的角度认识灾害发生、发展的过程。

Abstract：

Social media data are attracting an increasing number of attention for their high accessibility and effectiveness on indicating urban disasters. Studies and appliances about social media data are focusing on situational awareness and coding, disaster response and relief, damage assessment, etc. Hurricane Sandy, happened in 2012, becomes the second largest cyclone to hit the USA since 1900, which caused catastrophic damage to many areas especially New York City. Based on Twitter and concerning database, the research outlines the temporal and spatial characters of the information by coding schema development, tweet classification and spatial web portal analysis. The logit regression model in the study examines the explanatory power for varying demographic and socioeconomic variables. Miscalculation and error of using big data to reflect real situation are discussed within the scope of mass, material, access, and motivation (MMAM). Result shows that there is statistical significance between tweet number and population as well as landmarks. Demographic factors like education level, age, sex also influence tweet number. This study contributes to previous studies by profiling hurricane Sandy’s impacts using big data mining and analyzing.

王森，美国北卡罗来纳大学教堂山分校，硕士研究生。senwang@live.unc.edu
肖渝，博士，美国波特兰州立大学，副教授。yuxiao99@gmail.com
黄群英，博士，美国威斯康辛大学麦迪逊分校，副教授。qhuang46@wisc.edu
张纯（通信作者），博士，北京交通大学，副教授，副系主任。zhangc@bjtu.edu.cn

• [1] WATTS D, CEBRIAN M, ELLIOT M. Public response to alerts and warnings using social media: report of a workshop on current knowledge and research gaps[R]. National Academise Press, 2013.
  [2] HUANG Q, XIAO Y. Geographic situational awareness: mining tweets for disaster preparedness, emergency response, impact, and recovery[J]. ISPRS International Journal of Geo-Information, 2015, 4(3): 1549-1568.
  [3] 张洋, 吕斌, 张纯. 可持续城市防灾减灾与城市规划: 概念与国际经验[M]. 科学出版社, 2012.
  [4] SAGIROGLU S, SINANC D. Big Data: A Review[C] // Collaboration technologies and systems (CTS). 2013 International Conference on. IEEE, 2013: 42-47.
  [5] HRISTIDIS V, CHEN S C, Li T, et al. Survey of data management and analysis in disaster situations[J]. Journal of Systems and Software, 2010, 83(10): 1701-1714.
  [6] City of Los Angeles. Recovery and reconstruction plan. Emergency Operations Organizations, City of Los Angeles, California, 1994.
  [7] Department of Building and Safety. City of Los Angeles zoning code[Z]. (2005). http://www.lacity.org/ladbs/permits/codes.htm.
  [8] WILSON R C. The Loma Prieta Quake: what one city learned[M]. International City Management Association, 1991.
  [9] KUNREUTHER H. Incentives for mitigation investment and more effective risk management: the need for public–private partnerships[J]. Journal of Hazardous Materials, 2001, 86(1): 171-185.
  [10] 胡以志. 灾后重建规划理论与实践：以新奥尔良重建为例，兼论对汶川地震灾后重建的借鉴[J]. 国际城市规划, 2008(4): 66-70.

  [11] 张纯, 张洋, 吕斌. 唐山大地震后重建与恢复的反思：城市规划视角的启示[J]. 城市发展研究, 2012(5): 119-126.
  

  [12] MANYIKA J, CHUI M, BROWN B, et al. Big data: the next frontier for innovation, competition, and productivity[J]. Analytics, 2011.
  [13] SAVAGE M, DEVINE F, CUNNINGHAM N, et al. A new model of social class? findings from the BBC’s Great British class survey experiment[J]. Sociology, 2013, 47(2): 219-250.
  [14] TAPIA A H, LALONE N, KIM H W. Run amok: group crowd participation in identifying the Bomb and Bomber from the Boston Marathon Bombing[C]. Proceedings of the 11th ISCRAM, 2014.
  [15] MEIER P. How crisis mapping saved lives in Haiti[OL]. National Geographic, 2013. http://voices.nationalgeographic.org/2012/07/02/crisismapping-haiti/.
  [16] HOUSTON J B, HAWTHORNE J, PERREAULT M F, et al. Social media and disasters: a functional framework for social media use in disaster planning, response, and research[J]. Disasters, 2015, 39(1): 1-22.
  [17] LINDSAY B R. Social media and disasters: current uses, future options, and policy considerations[J]. Disasters, 2014, 39(1): 1-22.
  [18] CAMERON M A, POWER R, ROBINSON B, et al. Emergency situation awareness from Twitter for crisis management[C]. Proceedings of the 21st International Conference Companion on World Wide Web. ACM, 2012: 695-698.
  [19] SUTTON J, PALEN L, SHKLOVSKI I. Backchannels on the front lines: emergent uses of social media in the 2007 Southern California Wildfires[C]. Proceedings of the 5th International ISCRAM Conference. Washington DC, 2008: 624-632.
  [20] VIEWEG S E. Situational awareness in mass emergency: a behavioral and linguistic analysis of microblogged communications[J]. Proguest LIC, 2012: 300.
  [21] SAKAKI T, OKAZAKI M, MATSUO Y. Earthquake shakes Twitter users: real-time event detection by social sensors[C]. Proceedings of the 19th international conference on World wide web. ACM, 2010: 851-860. 
  [22] SIGNORINI A, SEGRE A M, POLGREEN P M. The use of Twitter to track levels of disease activity and public concern in the US during the influenza A H1N1 pandemic[J]. PLos One, 2011, 6(5): e19467.
  [23] KENT J D, CAPELLO JR H T. Spatial patterns and demographic indicators of effective social media content during the Horsethief Canyon Fire of 2012[J]. Cartography and Geographic Information Science, 2013, 40(2): 78-89. ttp://journals.plos.org/plosone/article?id=10.1371/journal.pone.0019467.
  [24] MANDEL B, CULOTTA A, BOULAHANIS J, et al. A demographic analysis of online sentiment during Hurricane Irene[C]. Proceedings of the Second Workshop on Language in Social Media. Association for Computational Linguistics, 2012: 27-36. http://cs.iit.edu/~culotta/pubs/mandel12demo.pdf.
  [25] KUMAR S, BARBIER G, ABBASI M A, et al. Tweettracker: an analysis tool for humanitarian and disaster relief[C]. ICWSM, 2011. http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.639.4272&rep=rep1&type=pdf.
  [26] GAO Huiji, BARBIER Geoffrey, GOOLSBY Rebecca. Harnessing the crowdsourcing power of social media for disaster relief[J]. IEEE Intelligent Systems 26, 2011(3): 10-14. http://ieeexplore.ieee.org/document/5898447/?reload=true.
  [27] ASHKTORAB Z A, BROWN C B, NANDI M C, et al. Tweedr: mining twitter to inform disaster response. ISCRAM, 2014. http://cs.iit.edu/~culotta/pubs/ashktorab14tweedr.pdf.
  [28] PUROHIT H, CASTILLO C, DIAZ F, et al. Emergency-relief Coordination on Social Media: Automatically Matching Resource Requests and Offers[J]. First Monday, 2013: 19. http://firstmonday.org/ojs/index.php/fm/article/view/4848.
  [29] LONGUEVILLE B D, LURASCHI G, SMITS P, et al. Citizens as sensors for natural hazards: a VGI integration workflow[J]. Geomatica, 2010, 64: 41-59. http://www.isprs.org/proceedings/XXXVIII/4-W13/ID_02.pdf.
  [30] SCHNEBELE E, CERVONE G. Improving remote sensing flood assessment using volunteered geographical data[J]. Natural Hazards & Earth System Sciences, 2013, 13(3): 669-677.
  [31] SCHNEBELE E, OXENDINE C, CERVONE G, et al. Using non-authoritative sources during emergencies in urban areas[M] // Computational approaches for urban environments. Springer International Publishing, 2015: 337-361.
  [32] XIAO Y, HUANG Q, WU K. Understanding social media data for disaster management[J]. Natural Hazards, 2015, 79(3): 1663-1679.
  [33] LINDA G, HOLLOWAY C. Hurricane Sandy after action: report and recommendations to Mayor Michael R. Bloomberg[R]. The City of New York, New York, NY 36, 2013.
  [34] OH O, KWON K H, RAO H R. An exploration of social media in extreme events: rumor theory and twitter during the Haiti Earthquake 2010[C]. International Conference on Information Systems, Icis 2010, Saint Louis, Missouri, Usa, December. DBLP, 2010: 231.

• 收起