Helping Crisis Responders Find the Informative Needle in the Tweet Haystack

Crisis responders are increasingly using social media, data and other digital sources of information to build a situational understanding of a crisis situation in order to design an effective response. However with the increased availability of such data, the challenge of identifying relevant information from it also increases. This paper presents a successful automatic approach to handling this problem. Messages are filtered for informativeness based on a definition of the concept drawn from prior research and crisis response experts. Informative messages are tagged for actionable data -- for example, people in need, threats to rescue efforts, changes in environment, and so on. In all, eight categories of actionability are identified. The two components -- informativeness and actionability classification -- are packaged together as an openly-available tool called Emina (Emergent Informativeness and Actionability).

[1]  Timothy Baldwin,et al.  langid.py: An Off-the-shelf Language Identification Tool , 2012, ACL.

[2]  Abby Stoddard,et al.  Haiti Earthquake Response: Context Analysis , 2010 .

[3]  Xiang Yao,et al.  The Design of a Dynamic Emergency Response Management Information System (DERMIS) , 2004 .

[4]  Kenny Meesters,et al.  Making sense of crises: the implications of information asymmetries for resilience and social justice in disaster-ridden communities , 2019 .

[5]  Muhammad Imran,et al.  Enabling Rapid Classification of Social Media Communications During Crises , 2016, Int. J. Inf. Syst. Crisis Response Manag..

[6]  Carlos Castillo,et al.  What to Expect When the Unexpected Happens: Social Media Communications Across Crises , 2015, CSCW.

[7]  Murel H. Folk.  Help , 1967 .

[8]  Xiang Zhang,et al.  Character-level Convolutional Networks for Text Classification , 2015, NIPS.

[9]  Robert Munro,et al.  Subword and Spatiotemporal Models for Identifying Actionable Information in Haitian Kreyol , 2011, CoNLL.

[10]  Kalina Bontcheva,et al.  Simple Open Stance Classification for Rumour Analysis , 2017, RANLP.

[11]  Brendan T. O'Connor,et al.  TweetMotif: Exploratory Search and Topic Summarization for Twitter , 2010, ICWSM.

[12]  Kalina Bontcheva,et al.  Corpus Annotation through Crowdsourcing: Towards Best Practice Guidelines , 2014, LREC.

[13]  Harith Alani,et al.  Semantic Wide and Deep Learning for Detecting Crisis-Information Categories on Social Media , 2017, SEMWEB.

[14]  Nima Khakzad,et al.  Hurricane Harvey Report: A fact-finding effort in the direct aftermath of Hurricane Harvey in the Greater Houston Region , 2017 .

[15]  Fernando Diaz,et al.  CrisisLex: A Lexicon for Collecting and Filtering Microblogged Communications in Crises , 2014, ICWSM.

[16]  Axel Schulz,et al.  Semantic Abstraction for generalization of tweet classification: An evaluation of incident-related tweets , 2016, Semantic Web.

[17]  Bartel Van de Walle,et al.  Coordination and information management in the Haiyan response: Observations from the field , 2014 .

[18]  Irina P. Temnikova,et al.  The Case for Readability of Crisis Communications in Social Media , 2015, WWW.

[19]  Kenny Meesters,et al.  A multidisciplinary perspective on supporting community disaster resilience in Nepal , 2016, ISCRAM.

[20]  Jeffrey Pennington,et al.  GloVe: Global Vectors for Word Representation , 2014, EMNLP.

[21]  Sarah Vieweg,et al.  Situational Awareness in Mass Emergency: A Behavioral and Linguistic Analysis of Microblogged Communications , 2012 .

[22]  Bartel Van de Walle,et al.  Humanitarian Information Management and Systems , 2008, Mobile Response.

[23]  Muhammad Imran,et al.  Twitter as a Lifeline: Human-annotated Twitter Corpora for NLP of Crisis-related Messages , 2016, LREC.

[24]  J. Harrald.  Agility and Discipline: Critical Success Factors for Disaster Response , 2006 .

[25]  Leysia Palen,et al.  Microblogging during two natural hazards events: what twitter may contribute to situational awareness , 2010, CHI.

[26]  Kalina Bontcheva,et al.  Microblog-genre noise and impact on semantic annotation accuracy , 2013, HT.

[27]  Fernando Diaz,et al.  Extracting information nuggets from disaster- Related messages in social media , 2013, ISCRAM.

[28]  Bartel Van de Walle,et al.  #Help. The Reality of Social Media Use in Crisis Response: Lessons from a Realistic Crisis Exercise , 2016, 2016 49th Hawaii International Conference on System Sciences (HICSS).

[29]  Shafiq R. Joty,et al.  Applications of Online Deep Learning for Crisis Response Using Social Media Information , 2016, ArXiv.

[30]  Carsten Eickhoff,et al.  Cognitive Biases in Crowdsourcing , 2018, WSDM.

[31]  Kalina Bontcheva,et al.  Broad Twitter Corpus: A Diverse Named Entity Recognition Resource , 2016, COLING.