When a disaster occurs, a large number of images and texts attached geographic information often flood the social network in the Internet quickly. All these information provide a new data source for timely awareness of disaster situations. However, due to the regional variation in the number of social media users and characteristics of information propagate in cyberspace, new problems arose in the pattern analysis of spatial point process represented by the check-in data, such as the correlation between check-in points density and disasters events density, the spatial relation between check-in points, the spatial heterogeneity of point pattern and associated influences. In this study, we took the No. 201614 Typhoon as an example and collected Sina Weibo data between September 14 and September 17, 2016 using keywords “Typhoon” and “Meranti”. We classified the Weibo texts using Support Vector Machine(SVM) algorithms, and constructed a disaster database containing relevant check-in information. In addition, considering the spatial heterogeneity of Weibo users, we proposed a weighted model based on user activity at the check-in points. Using Moran’s I of the global autocorrelation statistics, we compared the check-in data before and after adding weights and discovered obvious spatial autocorrelation of the check-in data in real geographical locations. We tested our model on Weibo data with keyword “rain” and “power failure”. The results show that series map generated by our model can reflect the typhoon disaster spatio-temporal process trends well.
A spatio-temporal process data model for characterizing marine disasters
