Typhoon Disaster Information Aggregation and Evolution Feature Expression with Social Media Based on Event Process Model
Real-time information of the process of typhoon events serves as important reference for disaster emergency.At the era of big data,microblog text has been gradual applied to the prevention,preparation,response and recovery of disaster manage-ment.However,previous studies mostly focused on the acquisition of different disaster information in microblog text,while ig-noring the structural integration of this fragmented information,and thus cannot reflect the dynamic process of typhoon events.This paper constructs a typhoon event information model considering the multi-granularity and dynamic characteristics of infor-mation from the spatio-temporal perspective.First,the multi-granularity representation model of the typhoon disaster event process is constructed by taking the object as the basic unit of the event and the state as the basis of the multi-scale measure-ment of the event process.Second,a typhoon event process mining method based on social media is proposed.Through the infor-mation aggregation method of multi-level semantic constraints,different state features of objects under specific spatio-temporal conditions are extracted.Besides,the evolution trajectories of different objects are analyzed based on the sequence of states con-nected in space-time,and the different evolution features of event processes are expressed from multi-dimensions.Third,taking the typhoon"Lekima"disaster event in 2019 as an example,the rationality of the multi-granularity typhoon disaster event process model and the effectiveness of the information aggregation method are verified in practical application.
typhoon disasterevent process modelinginformation aggregationsocial mediadisaster emergency response
万方数据
维普
