Social Media Alert and Response to Threats to Citizens (SMART-C) (Invited Paper) Nabil Adam US Department of Homeland Security Science & Technology Directorate Washington, DC, USA Jayan Eledath * , Sharad Mehrotra + , Nalini Venkatasubramanian + * SRI International, Princeton, NJ + University of California, Irvine Abstract— Social media, such as blogs, Twitter, and information portals, have emerged as the dominant communication mechanism of society. Exploiting such input to gain awareness of an incident is a critical direction for research in effective emergency management. In this paper we present an overview of the SMART-C system, which is part of the social media initiative at the Department of Homeland Security. The system aims to enable robust bidirectional communication between emergency management and the public at large throughout the disaster life- cycle via a multitude of devices and modalities including cell phones, MMS messages, text messages, blogs, Twitter, etc. A discussion of the major components of SMART-C and related research challenges is included. These components include mechanisms to model event level semantic information, a platform for implementing multi-sensor fusion, mechanisms for estimating the veracity of information, data cleaning to reduce uncertainty and enhance accuracy of event detection and notification, and spatiotemporal analyses for pattern and trend analyses for higher level observations. Keywords - social media; emergency management; alerting; robust data analytics I. INTRODUCTION Social media, such as blogs, Twitter, and information portals, have emerged as the dominant communication mechanism of today’s society. In the context of emergency management, exploiting such input to gain awareness of an incident is a critical direction for research. Dynamic real-time incident information collected from on-site human responders about the extent of damage, the evolution of the event, the needs of the community and the present ability of the responders to deal with the situation combined with information from the larger community could lead to more accurate and real time situational awareness that allows informed decisions, better resource allocation and thus a better response and outcome to the total crisis. Component technologies such as mobile devices, ubiquitous networking, location technologies, etc., which are integral components of a system that supports robust communication between response agencies and the public at large, have significantly matured to make such a system a reality. Social media and mobile devices are also changing how communities at large seek information in disasters. Reaching populations with customized and timely alerts through multiple channels (traditional media, Internet technologies, mobile services and social networks) can help inform those at risk and assure those not at risk with messages that accurately reflect the levels of vulnerability of the target population. Many past and current projects (e.g., [11]), as well as several studies conducted by the National Academies [14, 19] have indeed established the need for a robust, seamless, and scalable bidirectional communication between the response agencies and the public at large. A system that provides such a capability should support robust and scalable capabilities for real-time collection and processing of incident-related data from citizens using diverse media. Information obtained from social media and human input may be structured (e.g., interactive websites, speech and text dialog systems) or unstructured (e.g., textual reports, voice, video inputs, MMS messages). Components of such a system include mechanisms to model event-level semantic information, a platform for implementing multi-sensor fusion, mechanisms for estimating the veracity of information, data cleaning to reduce uncertainty and enhance accuracy of event detection and notification, and spatio-temporal analyses for pattern and trend analyses for higher level observations. In this paper we present such a system, Social Media Alert and Response to Threats to Citizens (SMART-C), which is part of the social media initiative at DHS [33]. SMART-C supports an extensible plug-n-play architecture into which new mechanisms, techniques, modalities, and systems can be further incorporated to both extend the set of supported functionalities and/or to leverage the functionalities provided in the context of specific systems. SMART-C’s capabilities include: • Enrichment of incident information, specifically, semantic enrichment through multi-modal analysis (text, speech, video) to create event level representation • Integration with other data sources such as demographics, socio-economic, environmental data. COLLABORATECOM 2012, October 14-17, Pittsburgh, United States Copyright © 2012 ICST DOI 10.4108/icst.collaboratecom.2012.250713