Disaster Hashtags in Social Media

ISPRS International Journal of Geo-Information

Nikita Murzintcev¹  Changxiu Cheng² 

[1] Institute of Geographic Sciences and Natural Resources Research, CAS, Beijing 100101, China;State Key Laboratory of Earth Surface Processes and Resource Ecology, Beijing Normal University, Beijing 100875, China;

关键词: twitter;    hashtags;    information retrieval;    disasters;   

DOI  :  10.3390/ijgi6070204

来源: DOAJ

【 摘 要 】

Social media is a rich data source for analyzing the social impact of hazard processes and human behavior in disaster situations; it is used by rescue agencies for coordination and by local governments for the distribution of official information. In this paper, we propose a method for data mining in Twitter to retrieve messages related to an event. We describe an automated process for the collection of hashtags highly related to the event and specific only to it. We compare our method with existing keyword-based methods and prove that hashtags are good markers for the separation of similar, simultaneous incidents; therefore, the retrieved messages have higher relevancy. The method uses disaster databases to find the location of an event and to estimate the impact area. The proposed method can also be adapted to retrieve messages about other types of events with a known location, such as riots, festivals and exhibitions.

【 授权许可】

Unknown