摘要
为快速制定救援行动方案,提出了一种多词共现分析方法将灾害应急任务从历史发生的灾害案例中提取出来的方法。该方法首先利用K-means聚类算法将预处理好的文本进行聚类分析,归纳出应急任务集。然后利用关联规则挖掘Apriori算法提取案例文本中的频繁项集作为共现词集,分析解释共现词集与应急任务之间的关系。最后,运用Gephi软件将共现词语网络可视化,进一步分析应急任务中要素之间的相关关系。以暴雨引发的地质灾害事故为例,对该方法进行了实验,结果表明该方法能有效的将暴雨-地质灾害事故案例文本中的应急任务提取出来,并通过网络可视化分析发现,在救援过程中,应急任务之间是联动配合的一个体系,协调好各个应急任务能缩短应急响应的时间,为制定或完善救援行动方案提供参考。
To quickly formulate rescue action plans,a method of extracting emergency tasks from historical disaster cases using multi-word co-occurrence analysis is proposed.Firstly,the K-means clustering algorithm is used to cluster the preprocessed text and summarize the emergency task set.Then,the Apriori algorithm of association rule mining is used to extract frequent itemsets from the case text as co-occurrence word sets,and analyze the relationship between co-occurrence word sets and emergency tasks is analyzed.Finally,the Gephi software is used to visualize the co-occurrence word network,and further analyze the correlation between the elements in the emergency tasks.Taking the geological disaster accident caused by rainstorm as an example,the method is experimentally verified.The results show that this method can effectively extract the emergency tasks from the text of rainstorm-geological disaster accident,and through network visualization analysis,it is found that in the rescue process,the emergency tasks are a coordinated system that work together,and coordinating various emergency tasks can shorten the emergency response time,and provide reference for formulating or improving rescue action plans.
作者
杨炼卿
许铭
马成龙
戴彩岩
赵红
龙昭岳
YANG Lianqing;XU Ming;MA Chenglong;DAI Caiyan;ZHAO Hong;LONG Zhaoyue(School of Engineering and Technology,China University of Geosciences(Beijing),Beijing 100083,China;Key Laboratory of Deep Geological Drilling Technology,Ministry of Land and Resources,China University of Geosciences(Beijing),Beijing 100083,China;Guizhou Institute of Labor Protection Science and Technology,Zunyi 563000,China)
出处
《灾害学》
CSCD
北大核心
2024年第2期208-211,234,共5页
Journal of Catastrophology
基金
贵州省科技计划项目“基于暴雨-崩滑-泥石流灾害链情景构建的救援推演关键技术研究与示范”(黔科合支撑[2021]一般513)。
关键词
暴雨-地质灾害
应急任务
文本挖掘
多词共现
关联规则挖掘
K-MEANS聚类算法
rainstorm-geological disaster
emergency plan
text mining
amulti-word co-occurrence
ssociation rules mining
K-means clustering algorithm
