Aid is Out There: Looking for Help from Tweets during a Large Scale Disaster

Proceedings of the 51st Annual Meeting of the Association for Computational Linguistics, pages 1619–1629, Sofia, Bulgaria, August 4-9 2013. c⃝2013 Association for Computational Linguistics Aid is Out There: Looking for Help from Tweets during a Large Scale Disaster Istv´an Varga† Motoki Sano† Kentaro Torisawa† Chikara Hashimoto† Kiyonori Ohtake† Takao Kawai§ Jong-Hoon Oh† Stijn De Saeger† †Information Analysis Laboratory, National Institute of Information and Communications Technology (NICT), Japan {istvan, msano, torisawa, ch, kiyonori.ohtake, rovellia, stijn}@nict.go.jp §Knowledge Discovery Research Laboratories, NEC Corporation, Japan t-kawai@bx.jp.nec.com Abstract The 2011 Great East Japan Earthquake caused a wide range of problems, and as countermeasures, many aid activities were carried out. Many of these problems and aid activities were reported via Twitter. However, most problem reports and corre- sponding aid messages were not success- fully exchanged between victims and lo- cal governments or humanitarian organi- zations, overwhelmed by the vast amount of information. As a result, victims could not receive necessary aid and humanitar- ian organizations wasted resources on re- dundant efforts. In this paper, we propose a method for discovering matches between problem reports and aid messages. Our system contributes to problem-solving in a large scale disaster situation by facilitat- ing communication between victims and humanitarian organizations. 1 Introduction The 2011 Great East Japan Earthquake in March 11, 2011 killed 15,883 people and destroyed over 260,000 households (National Police Agency of Japan, 2013). Accustomed way of living suddenly became unmanageable and people found them- selves in extreme conditions for months. Just after the disaster, many people used Twitter for posting problem reports and aid messages as it functioned while most communication channels suffered disruptions (Winn, 2011; Acar and Mu- raki, 2011; Sano et al., 2012). Examples of such problem reports and aid messages, translated from Japanese tweets, are given below (P1, A1). P1 My friend said infant formula is sold out. If somebody knows shops in Sendai-city where they still have it in stock, please let us know. A1 At Jusco supermarket in Sendai, you can still buy water and infant formula. If A1 would have been forwarded to the sender of P1, it could have helped since it would help the “friend” to obtain infant formula. But in re- ality, the majority of such reports/messages, es- pecially unforeseen ones went unnoticed amongst the mass of information (Ohtake et al., 2013). In addition, there were cases where many humani- tarian organizations responded to the same prob- lems and wasted precious resources. For instance, many volunteers responded to problems which were heavily reported by public media, leading to oversupply (Saijo, 2012). Such waste of re- sources could have been avoided if the organiza- tions would have successfully shared the aid mes- sages for the same problems. Such observations motivated this work. We de- veloped methods for recognizing problem reports and aid messages in tweets and finding proper matches between them. By browsing the discov- ered matches, victims can be assisted to over- come their problems, and humanitarian organiza- tions can avoid redundant relief efforts. We define problem …

Content truncated to 500 words