Research On Extraction And Aggregation Method Of Earthquake Disaster Toponym In Microblog

Since the 21st century,with the development of the Internet and the popularity of smart phones,social media platforms are constantly emerging and mature,.In recent years,social media platform has become the most rapid and convenient way for the general public to release and obtain disaster information in time.In this kind of social media data,there are a large number of words,pictures,videos and other information recording the current situation of the disaster area.As one of the representatives of social media platform,microblog provides a channel for the public to share information anytime and anywhere.It can become a bridge connecting information inside and outside the disaster area with its own advantages.But at the same time,the low threshold of microblog users,the large amount of data,unstructured characteristics,low credibility and fragmentation of information caused by the convenience of information release have become obstacles to the effective application of microblog data.Therefore,it is an urgent problem to extract and integrate effective disaster information which is useful to assist disaster emergency decision-making from the microblog data.Aiming at the current situation that the existing microblog data is seldom used in the field of earthquake emergency disaster and the use of microblog information is not complete,this paper proposes a set of extraction and aggregation method for microblog disaster information which is based on statistics and rules,focusing on the two key issues of accurate extraction and aggregation of microblog information:(1)Aiming at the accurate extraction of earthquake disaster toponym,this paper designs the algorithm of earthquake disaster toponym extraction.Firstly,the keyword table of earthquake disaster is set up according to the disaster information and the characteristics of earthquake disaster.STM model is used to screen the disaster text of microblog.Then,the Chinese toponym in the disaster text of microblog are identified by CRF model.Finally,the final disaster toponym is determined according to the rules for the use of multi class toponym.The algorithm can not only make full use of microblog data to accurately extract earthquake disaster toponym,but also effectively reduce the consumption of time and manpower in large-scale application.(2)According to the comprehensive understanding of earthquake disaster information,this paper analyzes the earthquake disaster information aggregation from three aspects: toponym aggregation,event aggregation and abnormal region identification,and show the aggregation results in the form of charts,maps and knowledge maps.The earthquake disaster aggregation based on toponym is the sorting and merging based on identifying the disaster toponym of single microblog data,and introducing the information credibility evaluation factor to modify the aggregation results;the earthquake disaster aggregation based on event analyzes a seismic disaster event from two dimensions of time and space;the anomaly area analysis algorithm is designed to identify the administrative units with abnormal information based on the aggregation results of toponym.This study enriches the aggregation method of earthquake disaster,and the results of aggregation analysis can assist the emergency management department in the overall understanding of earthquake disaster.(3)A platform for extracting and aggregating earthquake disaster toponym is designed and developed to verify the research content.The system is based on microblog data,according to the research results of this paper,the extraction and aggregation system of earthquake disaster place names is constructed.The functions of microblog data acquisition and preprocessing,disaster location extraction,disaster information aggregation and information visualization are realized.Taking Changning earthquake data as an example,the feasibility and reliability of the research content in this paper are verified.This method can obtain micro blog data related to events through disaster keywords,time,search area information and web results characteristics,and reduce the interference of unrelated data;through the construction of earthquake disaster knowledge base,it can make more comprehensive use of micro blog data,so as to obtain disaster location information;earthquake disaster information aggregation information,through charts and so on Methods,intuitively and completely display the spatial and temporal distribution of disaster events and the comparative relationship between the scope and degree of disaster.This paper makes full use of the effective disaster information in microblog,which can provide a certain auxiliary role for earthquake disaster emergency work.This method can obtain microblog data related to events through disaster keywords,time,search area information and web page result features,and reduce the interference of unrelated data.Through the construction of earthquake disaster keyword table,accurate disaster toponym information can be extracted from microblog data more comprehensively.The seismic disaster information aggregation technology directly shows the temporal and spatial distribution law of earthquake disaster events and identifies the abnormal areas through a variety of visualization methods.This paper makes full use of the effective disaster information in microblog,which can provide a certain auxiliary role for earthquake disaster emergency work.