LAUSR

Whenever a natural disaster occurs, a damage assessment must be conducted to determine the extent of the damage caused, in order to quickly and effectively undertake disaster response, recovery, and reconstruction efforts. It is important to consider not only natural phenomena, but the impact of the damage on local communities as well (which is a pressing concern at any disaster site). Although a conventional, field-survey-based disaster assessment can yield solid information, it still takes time to gauge the overall implications. While an SNS system can facilitate information collection in real time, it is riddled with problems such as unreliability, and the challenge of handling vast amounts of data. In this study we analyzed Twitter content that was generated after the 2018 Hokkaido Eastern Iburi Earthquake and was related to disaster response efforts at the site of the disaster, and used it to test an approach that combines and utilizes natural language processing and geo-informatics for disaster assessment. We then verified the use of this process in two different disaster response scenarios. In this paper, we discuss some possible approaches to disaster assessment that utilize SNS information analysis technology.