Disaster Information, Resilience, for Emergency and Crisis Technologies
Permanent URI for this collection
Browse
Recent Submissions
1 - 10 of 10
-
ItemSocial media processing in crisis response : an attempt to shift from data to information exploitation.( 2021-01-05)Information about the ongoing events is of the utmost importance during emergencies. Previous work in crisis informatics found new ways to pull data from unexploited sources, such as social media. But while the volume of information is crucial, the way the information is reported and provided becomes increasingly important as the volume grows. Yet, little has been achieved on information management. This article proposes a way to automatically organize information from social media data up to decision-makers. This organization is enabled by a metamodel \cite{benaben_metamodel_2016} designed to model crucial information in crises. The article is organized as follows. First, the organization of current social media processing systems is presented. Then, the article presents the metamodel used and how it is relevant to organized information in crisis events through the lens of the 6W's \cite{kropczynski_identifying_2018}. Finally, it walks through the implementation of the proposal based on the two previous parts.
-
ItemOpen Science at a time of the COVID-19 pandemic: a new opportunity to improve emergency response( 2021-01-05)The ongoing COVID-19 pandemic has become a major milestone encouraging a change from traditional scholarly communication practices and policies in favour of greater openness, sharing, and reuse. Interviews with South Korean and Australian experts has helped to highlight the factors that either enable or limit the impact of Open Science during a public health emergency, such as the COVID-19 outbreak. The paper categorised such factors as: contextual and external; institutional and regulatory; resource-based; individual and motivational, and supplemented this categorisation with the interviewees’ quotes to illustrate specific cases and examples. The institutional and regulatory factors are perceived as the most important ones by interviewees.
-
ItemMachine Learning for Readability Assessment and Text Simplification in Crisis Communication: A Systematic Review( 2021-01-05)In times of social media, crisis managers can interact with the citizens in a variety of ways. Since machine learning has already been used to classify messages from the population, the question is, whether such technologies can play a role in the creation of messages from crisis managers to the population. This paper focuses on an explorative research revolving around selected machine learning solutions for crisis communication. We present systematic literature reviews of readability assessment and text simplification. Our research suggests that readability assessment has the potential for an effective use in crisis communication, but there is a lack of sufficient training data. This also applies to text simplification, where an exact assessment is only partly possible due to unreliable or non-existent training data and validation measures.
-
ItemDesigning Chemical Emergency Response Systems Based on Open Data( 2021-01-05)Emergency situations call for reliable information to allow for an effective and timely response. While digitalization penetrates industry sectors at an increasing rate, advances in information technology are not fully leveraged in emergency situations today. Building upon the paradigm of openness in form of open data, we propose a solution that connects government-provided emergency services with private sector expertise and resources. We apply a design science research approach to design an artifact that provides information in real-time to fire departments and medical staff in case of chemical substance incidents. We showcase that by providing open data, private sector organizations acknowledge their responsibility to share critical data while creating value in the process. Our contribution in this paper is threefold: First, we derive design requirements for artifacts to address chemical substance incidents. Second, we design and evaluate an artifact to showcase its suitability. Third, we showcase value creation through open data.
-
ItemData Census of a Geographically-Bounded Tweet Set to Enhance Common Operational Picture Tools( 2021-01-05)Location information is of particular importance to crisis informatics. The Twitter API provides several methods to assess a rough location and/or the specific latitude and longitude in which a post originated. This paper offers a comparison of location information provided by Twitter’s four geolocation methods. The study aggregates one month of data from the greater Cincinnati, Ohio metropolitan area and assesses the relative contribution that each method can make to common operational picture tools used by crisis informatics researchers. Results show that of 49,744 Tweets, 4% contained geotags, 85.2% contained a location in the users’ profile, and 3.5% contained no apparent location data, but were gathered using the bounding box method and would not have been identified using traditional methods of gathering data using geotagged Tweets or user profile information alone. We reflect on these results in light of design implications for common operational picture tools (COPs).
-
ItemCrisis management and distrust: Study of an industrial accident in France( 2021-01-05)Most crises, whether ecological, humanitarian, economic or even social, are preceded by various warning signals which make it possible to trigger alerts. These alerts make it possible to prevent or limit human and material damage if they are delivered on time and if they allow stakeholders (decision-makers, populations, etc.) to prepare adequately for the coming crisis. Today, there are many systems that can pick up the warning signals of crises to limit their consequences. Warning systems are one of them, they have proven their effectiveness, but as with any system involving populations, a part of unpredictable remains. In this article, we are interested in industrial accidents and in particular the fire at the Lubrizol factory in Rouen (France), classified SEVESO, in September 2019. Following interviews with the population of the Rouen’s agglomeration, we set an example of distrust of public authorities’ decisions and give some leads on how an information system could reconcile populations and decision-makers.
-
ItemCitizens’ behaviours related to smoke in bushfires and their implications for computational models of evacuation( 2021-01-05)The behaviours of citizens during bushfires may determine whether they live or die. Using 100 citizen witness statements from the 2009 Australian bushfires, we show how people react to bushfire smoke. Eighty-nine witnesses expressly mention smoke, not necessarily in combination with fire. This prompted behaviours including: seeking further information, monitoring the situation, effecting a fire plan (including evacuation), alerting people to danger and fire risk, and going home. Computational simulators have been used to assess civilians’ risk and to help with evacuation efforts. Despite works that accurately model fire spread and people’s behaviours in response to perceiving fire, the issue of how people react to seeing smoke from a bushfire is rarely considered. We discuss how the identified behaviours may be incorporated into an agent-based simulator of bushfire.
-
ItemCharacterizing 311 System Reactions to a Global Health Emergency( 2021-01-05)Communities are complex, multi-dimensional systems that react to crises in a variety of different ways. Based on the municipal services provided to a community, 311 calls can be used as indicators of the different dimensions of that community’s reaction to a crisis situation. To improve Citizen Relationship Management, municipalities can analyze and even augment their 311 systems to capture specific types of information about an ongoing crisis. New York City did this by adding specific category types and descriptors to their 311 system, in response to the evolving COVID-19 pandemic. This paper provides an initial look at the 311 data for New York City and the variety of community behaviors that it is able to capture as a reaction to the pandemic and the associated actions taken by the authorities to respond to the situation.
-
ItemBuilding Community Resilience Using Gain-Loss Framing to Nudge Homeowner Mitigation and Insurance Decision-making( 2021-01-05)Promoting homeowners’ preparation for natural disasters is a critical component of building community resilience. Adoption of protective actions by individual homeowners could reduce the risk of injury and damage to property; however, despite extensive public education programs, numerous studies report that households still are under-prepared for natural disasters. The effectiveness of gain-loss framing to nudge risk averse decision-making has been demonstrated across several domains, yet the application of gain-loss framing effects for natural disaster preparation has concentrated only on policy-level decisions. A behavioral experiment (N= 1,840) was conducted to test whether gain-loss framing can be used to nudge homeowner risk mitigation and insurance purchase decisions. Consistent with Prospect Theory, results indicate that a gain-frame is more likely to lead to risk averse decisions to mitigate for floods and hurricanes, but not for earthquakes. Disaster specific framing effects for nudging individual mitigation decisions provide unique implications building community resilience.
-
Item