基于微博数据的地震灾情信息识别与应用  

Recognition and Application of Earthquake Disaster Information Based on Sin a Micro Blog Data

作　　者：孔珍 张叶廷[1] 杜志强[1] KONG Zhen;ZHANG Yeting;DU Zhiqiang(State Key Laboratory of Information Engineering in Surveying,Mapping and Remote Sensing,Wnhan University,Wuhan 430079,China)

机构地区：[1]武汉大学测绘遥感信息工程国家重点实验室,湖北武汉430079

出　　处：《地理信息世界》2020年第6期98-104,共7页Geomatics World

基　　金：国家重点研发计划项目(2017YFC1502902);国家自然科学基金项目(41971347,41871314)资助。

摘　　要：灾后的应急救灾初期急需受灾范围和受灾程度等灾情信息,为应急救援行动提供数据支撑。基于微博数据设计了地震灾情知识库,确定了灾情类别和灾情等级关键词;详细介绍了地震灾情知识库的构建过程及原理,首先采用基于种子词的方法对文本进行分类,然后利用条件随机场模型对文本中的中文地名进行识别,在通过多类地名使用规则确定地名信息后,使用灾情聚合计算方法得到地震的受灾区域和各区域对应的灾情指数,实现灾区行政地名的识别和灾情程度的比较,并以四川长宁地震为例进行了验证,结果表明,通过构建的地震灾情知识库和灾情聚合方法能够有效识别和整合互联网涉灾文本数据中的有效信息,辅助救援行动方案的制定。In the early stage of emergency relief after a disaster,information about the scope and extent of the disaster is urgently needed to provide data support for emergency rescue operations.Based on the Sina micro blog data,this paper designs the knowledge base of earthquake disaster,and determines the disaster category and disaster level keywords,and the construction process and principle of knowledge base of earthquake disaster are introduced in detail.Firstly the text is classified by using seed words.Then we use CRF model to identify Chinese toponym in the text.After the information of toponym is determined by using the rules,the disaster index of the affected area and each area are obtained by using the method of disaster aggregation calculation,so as to realize the identification of administrative place names in the disaster area and the comparison of disaster degrees.Taking Changning earthquake in Sichuan Province as an example,the results show that the knowledge base of earthquake disaster and the method of disaster aggregation constructed in this paper can effectively identify and integrate the effective information from the Internet data,and assist in the formulation of rescue action plan.

关 键 词：社交媒体涉灾数据 文本分类 中文地名识别 灾情信息聚合 

分 类 号：P315.1[天文地球—地震学]