DiReCT: Disaster Response Coordination with Trusted Volunteers

Mohammad Jahanian, Toru Hasegawa, Yoshinobu Kawabe, Yuki Koizumi, Amr Magdy, Masakatsu Nishigaki, Tetsushi Ohki, K.K.Ramakrishnan
Proceedings of 2019 International Conference on Information and Communication Technologies for Disaster Management (ICT-DM), pp.1-8, Dec.2019
[ Paper ] Best Paper Award


In disaster management, timely delivery of the right information to the right people can improve outcomes and save lives. With fewer first responders available, volunteers play an increasing role in disaster assistance. A communication framework that brings victims, volunteers, and first responders together for timely disaster assistance can be incredibly helpful. Recent observations, across multiple incidents, have shown that people extensively use social media for communicating in disaster situations. This paper proposes DiReCT, a framework for coordination of disaster response leveraging volunteers in a trustworthy fashion. DiReCT integrates social media with Information-Centric dissemination to help in disaster-related communications. To enable efficient dissemination, we exploit a dynamic namespace that captures the complex incident-management’s organizational relationships among participants’ roles, in addition to time and location. The DiReCT architecture facilitates both content retrieval and recipient-based publish/subscribe. Users, especially regular citizens and volunteers, may not have access to the full incident-management namespace. We propose a social media engine to intelligently map social media posts to the right name(s), using Natural Language Processing and Machine Learning, to steer social media posts towards appropriate first responder(s) that may be most appropriate in helping with the issue posted. Managing volunteers’ trust is crucial to ensure only credible posts get to the right people. We use a reputation-based trust model, that is used to guide the authorization, management and evaluation of trustworthiness of content and users. Preliminary evaluation results show the proposed social media and trust management-based solution can be effective.