Research

I have experienced significant research growth since promotion in five ways. First, I have seen steady growth in the number and quality of publications my team and I have published or presented. Second, I have seen growth in the number and size of the externally funded grants received. Third, I have built a strong international network of collaborators that participate in my research, exchange students and postdocs, and invite me to spend time in their laboratories. Fourth, I have been elected to central leadership positions in both of my home disciplines, and trusted to lead several international conferences. Lastly, I have received recognition in the forms of a Fulbright Scholar award, two Engaged Scholarship Awards, three nominations for Best Paper at a conference and a Best Journal Article award.

I am internationally known as a scholar of Crisis Informatics, the study of information discovery, needs, use and sharing in disaster or crisis settings. I have made three contributions in this space. First, I have identified unique network structures and barriers that exist in this form of technical crisis coordination in the disaster-relief sector at various levels of participating organizations. It is in these crisis coordination efforts that I find collaboration processes, projects and challenges that are both unique in their context and setting and similar to other ICT-based collaboration efforts. Influencing science, 53 publications and presentations have resulted from this work. Influencing practice, findings have spawned principles and best practices for over 300 Humanitarian Information Managers worldwide through a partnership with the United Nations Office of Coordination for Humanitarian Affairs (OCHA). These best practices were eventually adopted by the United Nations as part of OCHA’s Information Management Policies. The outcomes of this line of work received the Outstanding Paper Award in 2013 from the Emerald Literati Network.

Second, in partnership with the National Labs at Los Alamos and NASA’s Goddard Space Flight Center, I created Aurorasuarus, an early warning system and citizen science project (NSF-INSPIRE) that encourages public participation in improving space weather science through interest in the aurora borealis and which, most importantly, lead to implications for using crowdsourcing techniques for improving early warning systems in disaster response. The Aurora borealis was used as a proxy for a disaster in that its visibility was rare and unpredictable in mid latitudes. My work enabled crowdsourced aurora sightings to enhance scientific prediction of the effects of solar particles and energy upon the earth’s magnetosphere. Enhanced scientific prediction also served as an early warning system for auroral activity served back to participants through mobile notifications, thereby contributing to both research and citizen science communities. The practical crisis implications of this work produced six papers, one of which was nominated for a best paper award at the Conference for Information systems in Crisis Response and Management (ISCRAM).

Third, I contribute to the solution of one of the stickiest problems currently facing disaster response organizations—the organizational inability to take advantage of an abundance of citizen-produced social media data. My previous research has demonstrated the power of crowd-based action, identified people tweeting from the ground in mass disruption events, enhanced the role of community members through collaborative filtering of social media data, and collaboratively processed information in a crisis. Despite strong evidence as to the value of social media data during a crisis, there are numerous challenges to adoption of this data, including issues of reliability, verification, deception, quantity, and translation of reported observations to responders.  Manually sifting through voluminous streaming data to filter useful information in real time is inherently impossible. My team and I have shown promising results in using natural language processing, sentiment analysis, data mining and machine learning interventions in automating the process of identifying useful, relevant and trustworthy information in social media crisis data.

Through various projects (NSF-Big Data, NSF-CRISP, NSF-IIS-CHS) my research group and I have contributed to the scientific understanding of usage patterns of social media during a crisis, including communication among citizens, communication from authorities to citizens, from citizens to authorities; and among authorities, including inter-organizational social networks.  My findings are that affected citizens and helpers form emergent groups to overcome the emergency on-site; digital volunteers self-organize in virtual and technical communities to remotely support amateur and professional response; emergency services deploy incident management teams for professional on-site emergency response; and trusted digital volunteers are organized in Virtual Operations Support Teams to assist professional response in the virtual realm.

This work contributes to science in that it develops and deploys a reusable information technology infrastructure that classifies and aggregates tweets and text messages so that they can be more easily used by responding organizations. In addition, this advances science by developing techniques to automatically detect truth and deception within steams of social media data. I deploy these knowledge and techniques during a crisis in which data can be aggregated, processed and served to responders and the public in real time to make a difference in people and property protected. As a result, my work makes a direct contribution to practice by enhancing the ability of crisis responders to make use of crowdsourced data with a high level of confidence.

I am also an international scholar. In 2012 I was awarded a Fulbright where I partnered with the University of Costa Rica to teach Crisis Informatics and conduct Crisis Informatics research. In 2014 I hosted the International Association for Information Systems for Crisis Response and Management. I brought 300 scholars and practitioners from around the world to Penn State to discuss technologies to support communities in crisis. Later that year, I was awarded funding to study information brokers during the Ebola crisis in Africa, which resulted in the mapping of information flows for infectious disease responders.  In 2015, I partnered with the Organization for Tropical Studies (Central America) and the Xishuangbanna Tropical Botanical Garden (China) through funding from the National Science Foundation, to enable public participation in scientific efforts through mobile technologies. Recently, I have developed an International Research Partnership with Ecole Nationale Supérieure des Mines d’Albi-Carmaux, in Albi, France. The goals of this partnership are to create automated systems that monitor both human and physical sensors leading and improved situational awareness for emergency responders.