摘要
移动互联网和社交媒体中蕴含着大量有效信息,这些信息的利用已成为现代社会生活的重要部分。针对地震应急既有与发展需求,本文采用现有可收集的社交媒体信息,形成可实现的地震灾情获取与烈度快速评估方法。以新浪微博移动端为数据源,通过分布式爬虫技术有效获取了我国自2016年以来的11次破坏性地震及其灾害信息的相关文本数据,采用矩阵化及结构化处理实现综合考虑震情、民众情感等多因素耦合的数据集;通过机器学习中的人工神经网络模型建立了基于社交媒体数据的地震烈度快速评估方法。本研究对于震后的灾情快速获取和烈度快速评估具有一定的理论意义和重要的应用价值。
The mobile Internet and social media have become an important part of modern society,which contain a lot of effective information. To meet the requirements of earthquake emergency,this paper uses the existing social media data to form a feasible method of earthquake disaster acquisition and rapid assessment. With the real-time interactive information of micro-blog mobile as the data source,the 11 destructive earthquake related text data of China since 2016 are effectively obtained by distributed crawler technology. The multi factor coupling data set,such as the earthquake and the people’s emotion,is realized by matrix and structured processing. The fast evaluation method of seismic intensity based on social media data is realized by artificial neural network model. It has a certain theoretical significance and important application value for rapid acquisition of disaster and rapid evaluation of intensity after the earthquake.
作者
薄涛
李小军
陈苏
王玉婷
祁国良
BO Tao;LI Xiaojun;CHEN Su;WANG Yuting;QI Guoliang(Key Laboratory of Earthquake Engineering and Engineering Vibration,Institute of Engineering Mechanics,China Earthquake Administration,Harbin 150080,China;Beijing Earthquake Agency,Beijing 100080,China;Institute of Geophysics,China Earthquake Administration,Beijing 100081,China;Beijing Boliang Shenghe Teohnology Co.,Ltd.,Beijing 101214,China)
出处
《地震工程与工程振动》
CSCD
北大核心
2018年第5期206-215,共10页
Earthquake Engineering and Engineering Dynamics
基金
地震科技星火计划项目(XH19002
XH15001Y)
北京市自然科学基金项目(8164068)
中央级公益性科研院所重大研究计划专项(DQJB17C03)
地震应急青年重点任务项目(CEA_EDEM-201801)~~
关键词
社交媒体
地震烈度
快速评估
机器学习
人工神经网络
social media
seismic intensity rapid assessment
machine learning
artificial neural network
