Help by Any Means: Sociotechnical Protocols for Multimedia Requests for Assistance to Next-Generation 911 Services,
Funding Source: National Science Foundation, Number xxxx, Amount xxxx, NSF website: xxx
Leadership: Jessica Kropczynski (PI) with Rob Grace, Shane Halse and Andrea Tapia
Project Website: xxx
Abstract: This planning grant supports efforts to engage citizens, 911 dispatchers, and first responders to co-design Help by Any Means (HAM) protocols that will coordinate the use of multi-channel, multimedia requests for assistance during a crisis. These efforts will contribute to the transformation of emergency services around NG911 infrastructures, including the design of public service announcements educating citizens when and how to request help using alternative and multimedia 911 communications, as well as PSAP procedures for processing and integrating multimedia information in ways that enhance emergency response capacities. Additional contributions include understanding the social and technical implications of text-to-911 and social media-to-911 services as viewed by a range of stakeholders in the NG911 transition, as well as adduce insights into the design and management of data-driven public services beyond those of emergency response.
Emergency Management in the Time of Pandemic (COVID-19)
Funding Source: College of Information Sciences and Technology, Penn State University
Leadership: Andrea Tapia (PI) with Jamie Reep, Rob Grace, Shane Halse and Jess kropczynski
Date: May 2020 through May 2021
Project Website: xxx
Abstract: From a tactical emergency dispatch perspective, information flows from requestors, to call taker, on to emergency dispatch, first responders, and sometimes, back to the requestor. Information flow is more important during pandemics, which can overwhelm and disrupt emergency communications systems. Additionally, issues can impact global supply chain and logistics and impede emergency response. This research will develop silo-crossing technologies to address these issues in a pandemic setting and explores how the information sharing needs between emergency management and emergency dispatch are similar or different from other types of disasters. The goals of this research are as follows: To discover, illustrate and understand the patterns of human interaction during a pandemic through the proxy of social media. To contrast pandemic-related patterns of human interaction with those found surrounding other forms of disaster, e.g. hurricane, tornado, flooding and drought. To discover the information needs of Emergency Managers and Public Safety Answering Points (PSAPs—911 call centers) during a pandemic. To contrast these information needs with those found surrounding other forms of disaster, e.g. hurricane, tornado, flooding and drought. To collaboratively and iteratively re-design media data acquisition and display tools to allow Emergency Managers and PSAP dispatchers to make use of hyper-localized social media data during a crisis to serve their needs during a pandemic.
BIGDATA: IA: Collaborative Research: Domain Adaptation Approaches for Classifying Crisis Related Data in Social Media
Funding Source: National Science Foundation, BIGDATA 1741370, $400,000
Role: PI with C. Caragea (Co-PI) and D. Caragea (Co-PI)
Date: January 2018-December 2020
The project investigates the use of big-data analysis techniques for classifying crisis-related data in social media with respect to situational awareness categories, such as caution, advice, fatality, injury, and support, with the goal of helping emergency response teams identify useful information. A major challenge is the scale of the data, where millions of short messages are continuously posted during a disaster, and need to be analyzed. The use of current technologies based on automated machine learning is limited due to the lack of labeled data for an emergent target disaster, and the fact that every event is unique in terms of geography, culture, infrastructure, technology, and the people involved. To tackle the above challenges, domain adaptation techniques that make use of existing labeled data from prior disasters and unlabeled data from a current disaster are designed. The resulting models are continuously updated and improved based on feedback from crowdsourcing volunteers. The research will provide real, usable solutions to emergency response organizations and will enable these organizations to improve the speed, quality and efficiency of their response. The research provides novel solutions based on domain adaptation and deep neural networks to tackle the unique challenges in applying machine learning for crisis-related data analysis, specifically the volume and velocity challenges of big crisis data. Domain adaptation approaches enable the transfer of information from prior source disasters to an emergent target disaster. Deep learning approaches make it possible to employ large amounts of labeled source data and unlabeled target data, and to incrementally update the models as more labeled target data becomes available. Large-scale analysis across combinations of source and target crises will help identify patterns of transferable situational awareness knowledge. The resulting technical and social solutions will be blended together for use in data management and emergency response.
CHS: Small: Collaborative Research: Automating Relevance and Trust Detection in Social Media Data for Emergency Response
Funding Source: National Science Foundation, Cyber-Human Systems, 1526542 and 1526678, $250,000 https://www.nsf.gov/awardsearch/showAward?AWD_ID=1526542
Role: PI with A. Squicciarini (Co-PI) and C. Caragea (Co-PI)
Date: September 2015-August 2018
Abstract: The project will develop means to improve information quality and use in emergency response, increasing the value of using messaging and microblogged data from crowds of non-professional participants during disasters. Despite the evidence of strong value to those experiencing the disaster and those seeking information concerning the disaster, there has been very little effort in detecting the relevance and veracity of messages in social media streams. The problem of data verification is one of the largest problems confronting emergency-response organizations contemplating using social media data. This research directly addresses this known problem by methods to measure relevant and verifiable information. The results of this research will have a direct pipeline to organizations involved in emergency response. Therefore the research has the potential to help organizations, which respond to emergencies, make use of large amounts of citizen-produced data, which in turn may improve the speed, quality, and efficiency of emergency response leading to better support to those who need them, and more lives saved. This research will contribute to the field of Emergency and Disaster Studies by mapping the key decisions made during an emergency response, the information needs, type, form and flow during those decision points, and most importantly, assessing data quality and verifiable standards for each. It will also investigate relevant and verifiable identifiers (or features), provide weights, incorporate these into an analytical framework, and use the results of the analysis as input to scalable computational models. The work will design algorithms that can estimate the relevance and veracity of messages in a high-volume streaming text comprised of short messages.
CRISP Type 2/Collaborative Research: Resilience Analytics: A Data-Driven Approach for Enhanced Interdependent Network Resilience
Funding Source: National Science Foundation, CRISP (Critical Resilient Interdependent Infrastructure Systems and Processes) Division Of Civil, Mechanical, and Manufacturing, Infrastructure Management and Extreme Events. Number 1541155, $827,171
Role: PI with Kash Barker (Co-PI), James H Lambert (Co-PI), Laura McLay (Co-PI), Charles D Nicholson(Co-PI), Jose Emmanuel Ramirez-Marquez (Co-PI), C. Caragea (Co-PI), Chris Zobel (Co-PI)
Supplement CRISP Type 2/Collaborative Research: Resilience Analytics: A Data-Driven Approach for Enhanced Interdependent Network Resilience
Funding Source: National Science Foundation, CRISP (Critical Resilient Interdependent Infrastructure Systems and Processes) Division Of Civil, Mechanical, and Manufacturing, Infrastructure Management and Extreme Events. Number 1541155, $100,000
Role: PI with Jessica Kropczynski (Co-PI)
Date: May 2016- May 2017
Recent natural disasters have challenged our traditional approaches of planning for and managing disruptive events. Today, social media provides an opportunity to make use of community-driven data to help us understand the resilience, or lack thereof, of community networks (e.g., friends, neighborhoods) physical infrastructure networks (e.g., transportation, electric power) and networks of service providers (e.g., emergency responders, restoration crews). This Critical Resilient Interdependent Infrastructure Systems and Processes (CRISP) collaborative research integrates multiple disciplinary perspectives in engineering, computer science, and social science to address how community-driven data can help (i) understand the behavior of these interdependent networks before, during, and after disruptions, and (ii) more effectively reduce their vulnerability to and enhance their recovery after a disruption. The results will significantly improve our understanding and management of infrastructure recovery from natural disasters. Two research components comprise this effort in resilience analytics. The first component creates a network model of the interdependence of infrastructure networks, the community networks that they serve, and the service networks engaged to respond after a disruption. We will explore the functional relationships between community resilience and infrastructure network performance. Model results will enable decision makers to understand the balance of resilience across the several networks and regions. The second component integrates the interdependent network model with community-sourced data to develop a framework of data analytics to better understand and plan for resilience. This component builds on research in the field of socio-technical systems relating to the analysis of social media data monitored after a disruption. The methods will assess the value of information provided by crowd-sourced data with expertise of community social scientists. This project draws upon multiple methods across several disciplines. The multidisciplinary methods explored in this project are essential for a breakthrough in resilience analytics. This project aims at taking a significant step forward in our understanding of how real-time data from social media and other sources can describe, predict, and prescribe practices to manage interdependent networks in crises.
Scaling 911 Texting for Large-Scale Disasters: Developing Practical Technical Innovations for Emergency Management at Public Universities
Funding Source: College of IST—Seed Funding $10,000
Role: PI with Nikolas Giacobe (Co-PI)
Date: January 2015-June 2015
In a mass crisis event, Emergency Operation Centers EOC cannot meet the demand of thousands of individuals trying to alert or request emergency services. However, new technology, driven by the right policy and tested for strengths and weaknesses in a data rich, semi-predictable environment, can help to address current PSAP limitations. In this paper the authors present a system that aims to provide real-time data to emergency managers during a crisis event in such an environment-a college town during a football game or similarly attended event. The system is designed to accept, sort, triage and deliver hundreds of direct text messages from populations engaged in a crisis to emergency management staff who can respond. They posit that when a municipal or county-level EOC is cross-housed with a University EOC, multiple opportunities for development and funding occur. Universities can provide the technical expertise, funding, staffing, development and testing for systems that serve the EOC. Most importantly, Universities also provide disaster-like events that can be used as proxies for unpredictable mass crises during which more valid and reliable testing can occur. The authors present preliminary findings from a text-to-emergency service currently in use by Penn State University Athletics.
RAPID: Socio-technical systems and Big Data Analytics in the Ebola Response
Funding Source: National Science Foundation, Cyber-Human Systems (CHS) 1519023, $99,000
Role: Co-PI with C. Maitland (PI) and Vasant Honavar (Co-PI)
Date: December 2014-December 2015
In response to the 2014 Ebola virus outbreak in western Africa, dozens of humanitarian relief organizations as well as the CDC and U.S. military are providing medical assistance or logistics support to the relief effort. At the same time, a diverse range of volunteer technical communities (VTCs) and academics, as well as the humanitarian relief organizations themselves, are attempting to make use of “big data” to improve the response. These big data analyses are based on diverse data from diverse sources, including call records from mobile phone companies, health worker inventories from ministries, and daily case-data reports aggregated from multiple organizations. The analyses have generated outputs such as visually appealing maps and predictions of outbreak trajectories. However, precisely how, when and where these analyses can be used effectively by response organizations are still open questions. The project will develop knowledge to guide response organizations interested in leveraging existing and emerging big data from a variety of sources (response organizations, firms, government, individuals), which in turn may improve the speed, quality, and efficiency of crisis response. The research team of computer and social scientists will partner with a consultant with expertise in crisis information management deployed in the Ebola response. They will examine both organizational and technical dimensions of the use of big data analytics in the Ebola response organizations, carrying out a series of interviews to investigate how and where data is used (field, headquarters, or both) and the work involved to make big data analyses usable in the decision making of response organizations. The results will inform the development of a socio-technical systems framework to explain what makes big data analyses useable. The social dimensions of the framework will include the response context as well as decision making processes. The technical dimensions will include data availability, data analyses and output formats. In the process of developing this socio-technical framework, this research will identify mechanisms for matching organizational needs with big data analyses. More importantly, however, by identifying these mechanisms, the research will shed light on the fundamental roles of multi-level governance and articulation work in making effective use of big data analyses. The multi-level approach helps explain and predict the location in an organization?s hierarchy of technical and decision making expertise. Within these levels, a focus on articulation work helps specify the necessary tasks for using data in a highly dynamic environment. The project will extend the scholarship in the crisis informatics sub-field of big data beyond its current focus on social media data and provide clarification of “last mile” issues in big data aimed at ensuring the usefulness of output.
AURORASAURUS Citizen Scientists Experiencing the Extremes of Space Weather.
Funding Source: National Science Foundation Space Weather Research, Div Atmospheric and Geospace Sciences, INSPIRE. $998,957 https://www.nsf.gov/awardsearch/showAward?AWD_ID=1344296
Role: Co-PI, with Liz MacDonald (PI) and M. Hill (Co-PI)
Date: November 2013-November 2015
This is a two-year inter-disciplinary project pursuing tightly coupled goals within human centered computing, citizen science, and space weather research. The aurora borealis of the northern hemisphere and its twin, the aurora australis of the southern hemisphere, are among the most beautiful and awe-inspiring of natural phenomena. They are a manifestation of the interaction of solar plasma with the Earth’s atmosphere, magnetic field, and surface, the combined effect of which is termed space weather. As the aurora is a visible manifestation of space weather, observations of aurora are potentially a means of forecasting its catastrophic extremes. Capitalizing on public curiosity of normally intangible plasma physics, the objective of this project is to create a system for collecting, analyzing, interpreting, and redistributing data on the dynamics and evolution of auroral events using crowd-sourced ad hoc Tweets and more purposeful postings from citizen scientists. The current solar maximum is the first since the emergence of the ubiquitous use of social media that has changed – and will continue to change – our interactions with computers and the world. Building on a demonstrated prototype system, the project is poised to take advantage of the approach in 2013-2014 of the maximum in the current 11-year solar activity cycle, with several high activity years following. The team combines expertise in space weather science, human-computer interface design, and informal science education to realize each of its intertwined goals. This low-cost, citizen science system for improved forecasting of geomagnetic storms has the potential to transform the way space weather prediction is done and considering the enormous potential cost to society of damage due to such storms would be cost-effective. The project will help enhance public understanding of this little known phenomenon so that citizens are aware and prepared to respond to the effects of space weather. Resulting new understanding of effective approaches to citizen science and the impact of human computer interactions on motivations and success at learning will have value to a wealth of other ongoing citizen science programs.
EAGER: Collaborative Research: Establishing Trustworthy-Citizen-Created Data for Disaster Response and Humanitarian Action
Funding Source: National Science Foundation. Cyber-Human Systems (CHS) 1353400, $140.000
Role: PI. with A. Squicciarini (Co-PI) and Caragea, C.(Co-PI)
Date: August 2013-August 2014
Often referred to as microblogging, the practice of average citizens reporting on activities “on-the-ground” during a disaster is increasingly common. The contents of these message are potentially valuable to responder organizations and victims, but their volume makes it difficult to separate valuable messages from the stream. This project will examine microblogged messages sent during disasters to determine what aspects of the messages (individually and collectively) indicate that they are relevant, verifiable and actionable. Factors to be considered include the content of the messages, the identity of the sender and the overall pattern and spread of messages. The identified factors will then be used to instruct crowdsourced workers who will label messages to create a large corpus of labelled messages. The project is important because microblogging data are seen as increasingly important: they are ubiquitous, rapid and accessible, and they are believed to empower average citizens to become more situationally aware during disasters and to coordinate to help themselves. The result of the project, if it is successful, will be evidence that it is possible to identify relevant, verifiable and actionable messages from a stream of microblogged messages and identification of the evidentiary factors. A further outcome will be a disaster-related, labeled dataset of messages, which will be useful to researchers, e.g., those seeking to automatically classify information within a microblogged data stream.