The Emergence of Communities and Their Leaders on Twitter Following an Extreme Event

Abstract

Twitter is presently utilized as a channel of communication and information dissemination. At present, government and non-government emergency management organizations utilize Twitter to disseminate emergency relevant information. However, these organizations have limited ability to evaluate the Twitter communication in order to discover communication patterns, key players, and messages that are being propagated through Twitter regarding the event. More importantly there is a general lack of knowledge of who are the individuals or organizations that disseminate warning information, provide confirmations of an event and associated actions, and urge others to take action. This paper presents results of the analysis of two events—2011 Japan Tsunami and 2012 Hurricane Sandy. These results provide an insight into understanding human behavior, collectively as part of virtual communities on Twitter and individually as leaders and members of those communities. Specifically, their behavior is evaluated in terms of obtaining and propagating warning information, seeking and obtaining additional information and confirmations, and taking the prescribed action. The analysis will employ a methodology that shows how Natural Language Processing (NLP) and Social Network Analysis (SNA) can be integrated to provide these results. This methodology allows to extract actionable Twitter messages, construct actionable network, find actionable communities and their leaders, and determine the behaviors of the community members and their leaders. Moreover, the methodology identifies specific roles of the community leaders. Such roles include dispensing unique/new emergency relevant information, providing confirmations to the members of the communities, and urging them to take the prescribed action. The results show that the government agencies had limited participation on Twitter during 2011 Japan Tsunami compared to an extensive participation during 2012 Hurricane Sandy. The behavior of Twitter users during both events was consistent with the issuance of actionable information (i.e. warnings). The findings suggest higher cohesion among the virtual community members during 2011 Japan Tsunami than during 2012 Hurricane Sandy event. However, during both events members displayed an agreement on required protective action (i.e. if some members were propagating messages to take action the other members were taking action). Additionally, higher differentiation of leadership roles was demonstrated during 2012 Hurricane Sandy with stronger presence of official sources in leadership roles.