测绘学报 ›› 2023, Vol. 52 ›› Issue (3): 490-500.doi: 10.11947/j.AGCS.2023.20210496

• 地图学与地理信息 • 上一篇    下一篇

基于微博文本的灾害事件信息时空过程聚合方法

叶鹏1,2,3, 张雪英1,4, 张春菊5   

  1. 1. 南京师范大学虚拟地理环境教育部重点实验室, 江苏 南京 210023;
    2. 扬州大学城市规划与发展研究院, 江苏 扬州 225127;
    3. 扬州大学建筑科学与工程学院, 江苏 扬州 225127;
    4. 江苏省地理信息资源开发与利用协同创新中心, 江苏 南京 210023;
    5. 合肥工业大学土木与水利工程学院, 安徽 合肥 230009
  • 收稿日期:2021-08-27 修回日期:2022-06-15 发布日期:2023-04-07
  • 通讯作者: 张雪英 E-mail:zhangsnowy@163.com
  • 作者简介:叶鹏(1991-),男,博士,讲师,研究方向为地理大数据挖掘。E-mail:007839@yzu.edu.cn
  • 基金资助:
    国家自然科学基金(41971337);国家重点研发计划(2021YFB3900903);江苏高校哲学社会科学研究一般项目(2022SJYB2128);虚拟地理环境教育部重点实验室开放课题(2021VGE01)

Spatio-temporal process based information aggregation method of disaster events in microblog text

YE Peng1,2,3, ZHANG Xueying1,4, ZHANG Chunju5   

  1. 1. Key Laboratory of Virtual Geographic Environment (Nanjing Normal University), Ministry of Education, Nanjing 210023, China;
    2. Urban Planning and Development Institute, Yangzhou University, Yangzhou 225127, China;
    3. College of Architectural Science and Engineering, Yangzhou University, Yangzhou 225127, China;
    4. Jiangsu Center for Collaborative Innovation in Geographical Information Resource Development and Application, Nanjing 210023, China;
    5. School of Civil Engineering, Hefei University of Technology, Hefei 230009, China
  • Received:2021-08-27 Revised:2022-06-15 Published:2023-04-07
  • Supported by:
    The National Natural Science Foundation of China (No. 41971337);The National Key Research and Development Program of China (No. 2021YFB3900903);The Key Project of Philosophy and Social Science Research in Colleges and Universities in Jiangsu Province (No. 2022SJYB2128);The Open Fundation of Key Laboratory of Virtual Geographic Environment (Nanjing Normal University), Ministry of Education (No. 2021VGE01)

摘要: 各类灾害事件频发已成为全球可持续发展面临的重大威胁。在大数据环境下,微博文本逐渐被应用于灾害管理的预防、准备、响应和恢复工作。以往研究多关注微博文本中灾情信息的获取,却忽略对这些碎片化信息进行有序化整合。本文从时空视角构建多层次的灾害事件信息模型,在抽取出微博文本中灾害事件信息要素的基础上,提出基于“对象-状态”的过程信息聚合方法,解决微博文本中灾害事件信息分散化、时空粒度多样化和无序化的问题。基于新浪微博进行台风“利奇马”事件的案例分析,结果表明,本文方法能够全面地获取灾害事件过程中各个时空节点上的灾情信息,有利于从微博文本中挖掘小尺度下的灾害突发状况。

关键词: 灾害事件, 时空过程, 多层次信息聚合, 微博文本

Abstract: The frequent occurrence of various disaster events has become a major threat to global sustainable development. In the era of big data, microblog text has been gradually applied to the prevention, preparation, response and recovery of disaster management. However, previous studies mostly focused on the acquisition of different disaster information in microblog text, while ignoring the structural integration of this fragmented information. In this paper, an information aggregation method of disaster events in microblog text is proposed. First, a multi-level disaster event information model is constructed from the spatio-temporal perspective. Then, the information elements of disaster events are extracted from the microblog text. Finally, a process-oriented information aggregation model based on “Object-State” is constructed. On this basis, it can be solved that the problems of decentralization, spatio-temporal granularity diversification and disorder of disaster event information in microblog text. Based on the case study of typhoon “Lekima” event using Sina Weibo, the results show that the method proposed in this paper can comprehensively acquire the information of different objects on every spatio-temporal node during the process of typhoon events, which is beneficial to mining disaster emergencies in small scale from microblog text.

Key words: disaster events, spatio-temporal process, multi-level information aggregation, microblog text

中图分类号: