Shubham Gupta, & Craig A. Knoblock. (2010). Building geospatial mashups to visualize information for crisis management. In C. Zobel B. T. S. French (Ed.), ISCRAM 2010 – 7th International Conference on Information Systems for Crisis Response and Management: Defining Crisis Management 3.0, Proceedings. Seattle, WA: Information Systems for Crisis Response and Management, ISCRAM.
Abstract: In time-sensitive environments such as disaster management, decision-making often requires rapidly gathering the information from diverse data sources and then visualizing the collected information to understand it. Thus, it is critical to reduce the overhead in data integration and visualization for efficient decision-making. Geospatial mashups can be an effective solution in such environments by providing an integrated approach to extract, integrate and view diverse information. Currently, mashup building tools exist for creating mashups, but none of them deal with the issue of data visualization. An improper visualization of the data could result in users wasting precious time to understand the data. In this paper, we introduce a programming-by-demonstration approach to data visualization in geospatial mashups that allows the users to customize the data visualization.
|
Henning Gösling, & Jutta Geldermann. (2014). Methodological tool kit for humanitarian logistics. In and P.C. Shih. L. Plotnick M. S. P. S.R. Hiltz (Ed.), ISCRAM 2014 Conference Proceedings – 11th International Conference on Information Systems for Crisis Response and Management (pp. 190–194). University Park, PA: The Pennsylvania State University.
Abstract: Disasters trigger the need for relief items. The flows of relief items to the beneficiaries in the disaster area are determined by humanitarian logistics networks. The setup and operations of such networks can be supported by employing Operations Research models. Several models, based on different methodologies are available to support decision-making in the field of humanitarian logistics. This work-in-progress analysis suggests a structure of a methodological tool kit for humanitarian logistics. With its help, practitioners in the field of humanitarian logistics should be better equipped to find, compare, and apply available analytical models for their individual decision problem. To serve as an illustration, one OR model is presented in detail according to the proposed structure of the methodological tool kit.
|
Hafiz Budi Firmansyah, Jesus Cerquides, & Jose Luis Fernandez-Marquez. (2022). Ensemble Learning for the Classification of Social Media Data in Disaster Response. In Rob Grace, & Hossein Baharmand (Eds.), ISCRAM 2022 Conference Proceedings – 19th International Conference on Information Systems for Crisis Response and Management (pp. 710–718). Tarbes, France.
Abstract: Social media generates large amounts of almost real-time data which has proven valuable in disaster response. Specially for providing information within the first 48 hours after a disaster occurs. However, this potential is poorly exploited in operational environments due to the challenges of curating social media data. This work builds on top of the latest research on automatic classification of social media content, proposing the use of ensemble learning to help in the classification of social media images for disaster response. Ensemble methods use multiple learning algorithms to obtain better predictive performance than could be obtained from any of the constituent learning algorithms alone. Experimental results show that ensemble learning is a valuable technology for the analysis of social media images for disaster response,and could potentially ease the integration of social media data within an operational environment.
|
Daniel Hahn. (2007). Non-restrictive linking in wireless sensor networks for industrial risk management. In K. Nieuwenhuis P. B. B. Van de Walle (Ed.), Intelligent Human Computer Systems for Crisis Response and Management, ISCRAM 2007 Academic Proceedings Papers (pp. 605–609). Delft: Information Systems for Crisis Response and Management, ISCRAM.
Abstract: The OSIRIS project addresses the disaster management workflow in the phases of risk monitoring and crisis management. Risk monitoring allows the continuous observation of endangered areas combined with sensor deployment strategies. The crisis management focuses on particular events and the support by sensor networks. Four complementary live demonstrations will validate the OSIRIS approach. These demonstrations include water contamination, air pollution, south European forest fire, and industrial risk monitoring. This paper focuses on the latter scenario: the industrial risk monitoring. This scenario offers the special opportunity to demonstrate the relevance of OSIRIS by covering all the aspects of monitoring, preparation and response phases of both environmental risk and crisis management. The approach focuses on non-restrictive linking in a wireless sensor network in order to facilitate the addition and removal of nodes providing open interaction primitives allowing the comfortable integration, exclusion, and modification. A management layer with an event-triggered and service-based middleware is proposed. A live lab with real fire is illustrated.
|
Haiyan Hao, & Yan Wang. (2020). Hurricane Damage Assessment with Multi-, Crowd-Sourced Image Data: A Case Study of Hurricane Irma in the City of Miami. In Amanda Hughes, Fiona McNeill, & Christopher W. Zobel (Eds.), ISCRAM 2020 Conference Proceedings – 17th International Conference on Information Systems for Crisis Response and Management (pp. 825–837). Blacksburg, VA (USA): Virginia Tech.
Abstract: The massive crowdsourced data generated on social networking platforms (e.g. Twitter and Flickr) provide free, real-time data for damage assessment (DA) even during catastrophes. Recent studies leveraging crowdsourced data for DA mainly focused on analyzing textual formats. Crowdsourced images can provide rich and objective information about damage conditions, however, are rarely researched for DA purposes. The highly-varied content and loosely-defined damage forms make it difficult to process and analyze the crowdsourced images. To address this problem, we propose a data-driven DA method based on multi-, crowd-sourced images, which includes five machine learning classifiers organized in a hierarchical structure. The method is validated with a case study investigating the damage condition of the City of Miami caused by Hurricane Irma. The outcome is then compared with a metric derived from NFIP insurance claims data. The proposed method offers a resource for rapid DA that supplements conventional DA methods.
|
Niklas Hallberg, Helena Granlund, Jonas Hallberg, & Rogier Woltjer. (2012). Rationale for emergency management systems for local communities: A needs assessment. In Z.Franco J. R. L. Rothkrantz (Ed.), ISCRAM 2012 Conference Proceedings – 9th International Conference on Information Systems for Crisis Response and Management. Vancouver, BC: Simon Fraser University.
Abstract: Information systems have great potential to support emergency management. However, development of such systems is difficult, due to the complexity of emergency management. The ability to be able to reveal the needs for support are essential for successful system developments. The emergency responders at the local community level are the main actors when it comes to emergency management. The objective of this paper is to explore the rationale for emergency management systems at the local community level. This is done by an extensive needs assessment based on 12 interviews with representatives for local as well as regional emergency organizations and 49 governing documents. The analysis uncovers ten areas where emergency management systems can enhance the ability of local communities' to manage emergencies. © 2012 ISCRAM.
|
Stephen E. Hannestad. (2005). Incident command system: A developing national standard of incident management in the U.S. In B. C. B. Van de Walle (Ed.), Proceedings of ISCRAM 2005 – 2nd International Conference on Information Systems for Crisis Response and Management (pp. 19–28). Brussels: Royal Flemish Academy of Belgium.
Abstract: This paper is a critical examination of the U.S. Incident Command System (“ICS”) as an organizational structure and information gathering tool for emergency management from the perspective of a career emergency management officer who became a graduate student in information management following 35 years of public service. In addition to examining the ICS, and assessing its current weaknesses in the area of information management, the paper proposes a low-cost, COTS approach to automating the ICS information gathering and dissemination process.
|
|
Hans Betke. (2015). Structure and Elements of Disaster Response Processes ? A General Meta-Model. In L. Palen, M. Buscher, T. Comes, & A. Hughes (Eds.), ISCRAM 2015 Conference Proceedings ? 12th International Conference on Information Systems for Crisis Response and Management. Kristiansand, Norway: University of Agder (UiA).
|
John R. Harrald, Theresa I. Jefferson, Frank Fiedrich, Sebnem Sener, & Clinton Mixted-Freeman. (2007). A first step in decision support tools for humanitarian assistance during catastrophic disasters: Modeling hazard generated needs. In K. Nieuwenhuis P. B. B. Van de Walle (Ed.), Intelligent Human Computer Systems for Crisis Response and Management, ISCRAM 2007 Academic Proceedings Papers (pp. 51–56). Delft: Information Systems for Crisis Response and Management, ISCRAM.
Abstract: The US has not yet developed adequate models for estimating hazard generated needs, the necessary first step for developing useful decision support systems needed to estimate the capability and capacity of the response forces required. Modeling and technology required to support the decisions made by humanitarian relief organizations requires scenario driven catastrophic planning. This paper documents the lack of effective decision support tools and systems for humanitarian aid and describes the current state of models and methods used for determination of hazard generated needs. The paper discusses work performed on a catastrophic earthquake preparedness project. It outlines how the results of this project will be used to advance the modeling and decision support capabilities of federal, state and local disaster planners and emergency responders.
|
Alison J. Hayes, Jessica Lancaster, Zeno Franco, & Anne Kissack. (2012). Disaster medical education & simulated crisis events: A translational approach. In Z.Franco J. R. L. Rothkrantz (Ed.), ISCRAM 2012 Conference Proceedings – 9th International Conference on Information Systems for Crisis Response and Management. Vancouver, BC: Simon Fraser University.
Abstract: This review addresses current educational and research efforts in disaster medical education (DME) in the United States. Since the events of 9/11, DME has received greater attention. However substantial problems remain in terms of ensuring that large numbers of medical students and residents are exposed to high quality DME – not only Emergency Medicine residents. Barriers to widespread adoption of DME include lack of performance metrics, disagreement task areas, and lack of emphasis on physician leadership. Further, such efforts must ensure retention of key information over periods that are disaster free; utilize objective training metrics that will allow for an evidence base to form; and develop low cost, scalable training approaches that offer greater fidelity to the disaster environment than classroom based instruction. To improve the state of the art, we argue that DME research must move toward a translational science model that integrates important advances in basic information science into application that improve the clinical performance of frontline medical staff who are called on to respond to individual and community needs in the aftermath of disaster. Mid-fidelity, team-in-the-loop simulations developed for disaster manager training may provide an avenue toward improved DME by exposing medical students to scenarios that fundamentally challenge their assumptions in real-time game play. This can be accomplished with lower costs and greater scalability than live exercise or mock-up training approaches. © 2012 ISCRAM.
|
Arthur H. Hendela, Murray Turoff, & Starr Roxanne Hiltz. (2010). Cross impact security analysis using the HACKING Game. In C. Zobel B. T. S. French (Ed.), ISCRAM 2010 – 7th International Conference on Information Systems for Crisis Response and Management: Defining Crisis Management 3.0, Proceedings. Seattle, WA: Information Systems for Crisis Response and Management, ISCRAM.
Abstract: Security of network assets is a high priority with little traditional return on investment. Increasingly, cyber attacks are being used by both terrorist and unfriendly government organizations. The HACKING Game, a Cross Impact Analysis planning tool, can be used to plan security resource allocation in computer networks. Cross Impact Analysis provides a mathematical basis to determine the interrelationships of one event with a set of other events. Output from the HACKING Game's Cross Impact Analysis model can be used to help justify security expenditures, with an added benefit of being a training tool for employees learning to protect networks. This paper presents details of the Hacking Game's design and its capabilities. Cross impact modeling can be used to develop games for any situation characterized by a set of offense and defense events to produce an individual or collaborative model for such things as natural and man-made disasters.
|
Henke, S., Widera, A., & Hellingrath, B. (2023). Evaluation-driven Disaster Management Exercises: A Collaborative Toolkit. In Jaziar Radianti, Ioannis Dokas, Nicolas Lalone, & Deepak Khazanchi (Eds.), Proceedings of the 20th International ISCRAM Conference (pp. 989–999). Omaha, USA: University of Nebraska at Omaha.
Abstract: Disaster management exercises are a core component of humanitarian organizations’ preparedness strategies. They entail diverse purposes, from training capabilities of participants to testing response plans to enhancing collaboration between organizations and many more. However, it is uncertain how much exercises contribute to preparedness. Rigorous evaluation is needed to exploit learning opportunities of an exercise. Therefore, exercises must target evaluable objectives, which is complicated by the socio-technical openness of the exercise system, the heterogeneity of organizational needs, and the scarcity of resources. Many different tools aim to support evaluation but are limited to specific use cases, resulting in a fragmented overview for practitioners. Due to the excessive effort involved, practitioners often consider exercise evaluation to be of secondary importance. This study thus proposes the conceptual design of a combined toolkit that supports the practitioners in a more rigorous but resource-efficient evaluation to make disaster management exercises more evaluation-driven.
|
Sergio Herranz, David Díez, Díaz, P., & Starr Roxanne Hiltz. (2012). Exploring the design of technological platformsfor virtual communities of practice. In Z.Franco J. R. L. Rothkrantz (Ed.), ISCRAM 2012 Conference Proceedings – 9th International Conference on Information Systems for Crisis Response and Management. Vancouver, BC: Simon Fraser University.
Abstract: Virtual Communities of Practice (VCoP) refers to groups of people who share a concern about a specific domain or topic and use a virtual environment to share and increase their knowledge and expertise about this domain. This kind of social structure has intrinsic features suitable to support emergency management communities. Nevertheless, the design of specific technological platforms that support both the activity and the practice of the community is not a trivial task, especially in critical domains such as emergency management. This paper presents the inquiry process carried out over one and a half years for the purpose of generating insights about the application of VCoPs within the emergency management context. Based on a case study, a set of findings is presented about the guidelines that should be followed in order to develop suitable technological platforms that support the labor of VCoPs in the emergency management context. © 2012 ISCRAM.
|
Norman C. Hester, Jim Wilkinson, Stephen Patrick Horton, & Theresa I. Jefferson. (2008). Integration of information systems for post earthquake research response. In B. V. de W. F. Fiedrich (Ed.), Proceedings of ISCRAM 2008 – 5th International Conference on Information Systems for Crisis Response and Management (pp. 362–367). Washington, DC: Information Systems for Crisis Response and Management, ISCRAM.
Abstract: Natural disasters occur infrequently, limiting our ability to develop an inclusive knowledge base concerning such events. The ability to study, interpret, and document findings immediately following a damaging seismic event, is a critical step in furthering our understanding of events, allowing for effective awareness, mitigation, response, and recovery efforts. In the central United States, a Post Earthquake Technical Information Clearinghouse (PETIC) Plan has been developed to coordinate research activities, and to facilitate collaboration between the emergency management and research communities. Because a damaging earthquake in the central U.S. will impact several states, a Multi-State Technical Information Clearinghouse (MSTIC) Coordination Plan to link state technical information clearinghouses (STICs) is proposed. This paper describes beginning efforts to define the role and functions of a MSTIC as well as formalize plans with emergency management agencies to facilitate collaboration and coordination between STICs, the MSTIC.
|
Erich Heumüller, Sebastian Richter, & Ulrike Lechner. (2012). Towards a framework for command post exercises. In Z.Franco J. R. L. Rothkrantz (Ed.), ISCRAM 2012 Conference Proceedings – 9th International Conference on Information Systems for Crisis Response and Management. Vancouver, BC: Simon Fraser University.
Abstract: The paper describes work-in-progress of an action research approach investigating command post exercises in disaster response organizations. The empirical basis of our approach is three command post exercises in Germa-ny. The main contribution of this paper is the framework for the conceptualization of command post exercises and an analysis of challenges in the conceptualization of command post exercises. © 2012 ISCRAM.
|
Michael Hiete, & Mirjam Merz. (2009). An indicator framework to assess the vulnerability of industrial sectors against indirect disaster losses. In S. J. J. Landgren (Ed.), ISCRAM 2009 – 6th International Conference on Information Systems for Crisis Response and Management: Boundary Spanning Initiatives and New Perspectives. Gothenburg: Information Systems for Crisis Response and Management, ISCRAM.
Abstract: Natural and man-made hazards may affect industrial production sites by both direct losses (due to physical damage to assets and buildings) and indirect losses (production losses). Indirect losses, e.g. from production downtimes, can exceed direct losses multiple times. Thus, the vulnerability of industrial sectors to indirect losses is an important component of risk and its determination is an important part within risk analysis. In this paper a conceptual indicator framework is presented which allows to assess the indirect vulnerability of industrial sectors to different types of disasters in a quantitative manner. The results are useful for information sharing and decision making in crisis management and emergency planning (mitigation measures, business continuity planning), since the developed indicator system helps to take the complex phenomenon of industrial vulnerability and the underlying interdependencies into account. Besides the identification and conceptual motivation of the indicators, methodical aspects such as standardization, weighting and aggregation are addressed.
|
Ehren Hill, & Frank Hardisty. (2012). CR-Site: An infrastructure siting tool for crisis response. In Z.Franco J. R. L. Rothkrantz (Ed.), ISCRAM 2012 Conference Proceedings – 9th International Conference on Information Systems for Crisis Response and Management. Vancouver, BC: Simon Fraser University.
Abstract: Many crisis response and recovery efforts require choosing locations in order to deliver needed materials and services. Automated methods can help choose optimal locations for relief camps, field hospitals, command centers, and other critical relief infrastructure. However, current information technology tools for siting relief infrastructure suffer from exposing too much complexity to the user. We are developing a tool, CR-Site, which we hope will serve as an exemplar of an emergency siting tool that eliminates unnecessary complexity, while exposing necessary parameters. In this paper, we describe the technical design and user workflow for CR-Site and provide a case study for the functionality provided by CR-Site. © 2012 ISCRAM.
|
Starr Roxanne Hiltz, & Linda Plotnick. (2013). Dealing with information overload when using social media for emergency management: Emerging solutions. In J. Geldermann and T. Müller S. Fortier F. F. T. Comes (Ed.), ISCRAM 2013 Conference Proceedings – 10th International Conference on Information Systems for Crisis Response and Management (pp. 823–827). KIT; Baden-Baden: Karlsruher Institut fur Technologie.
Abstract: Several recent studies point the way to enabling emergency response managers to be able to find relevant posts and incorporate them into their sensemaking and decision making processes. Among the approaches that have improved the ability to find the most relevant information are the social conventions of creating topic groups and tags and of “retweeting;” the use of trained volunteers to filter and summarize posts for responders; automated notifications of trending topics; natural language processing of posts; techniques for identifying posts from the disaster site; and the use of GIS and crisis maps to visually represent the distribution of incidents.
|
Starr Roxanne Hiltz, Jose J. Gonzalez, & Murray Turoff. (2013). ICT support and the effectiveness of decision making in disasters: A preliminary system dynamics model. In J. Geldermann and T. Müller S. Fortier F. F. T. Comes (Ed.), ISCRAM 2013 Conference Proceedings – 10th International Conference on Information Systems for Crisis Response and Management (pp. 668–673). KIT; Baden-Baden: Karlsruher Institut fur Technologie.
Abstract: A high level conceptual model is presented of factors hypothesized to be key determinants of the effectiveness of decision making in large scale disasters, grounded in the literature on disaster management. ICT robustness (including the use of social media) sensemaking, and the effectiveness of decision making processes by the multi-organizational Partially Distributed Teams that must cooperate are accorded key roles in the process model. The outcomes of the decision making processes modeled are decisions, in terms of timeliness and quality.
|
Starr Roxanne Hiltz, Linda Plotnick, & Rosalie J. Ocker. (2011). Learning about Emergency Management Information Systems through Partially Distributed Team projects. In E. Portela L. S. M.A. Santos (Ed.), 8th International Conference on Information Systems for Crisis Response and Management: From Early-Warning Systems to Preparedness and Training, ISCRAM 2011. Lisbon: Information Systems for Crisis Response and Management, ISCRAM.
Abstract: What are effective ways to train students for designing Emergency Management Information Systems and for working with distant partners on EMIS projects? This paper describes the results of a project whose goal is to assess the effectiveness of Partially Distributed Teams (PDTs) as a pedagogical strategy. A PDT is one type of distributed team, in which two or more co-located subgroups use information and communication technology to collaborate on a task. Much of emergency management involves cross-organizational and often cross-national collaboration in this form. Students need, but lack, learning experiences focused on how to work effectively in such situations. This field study involved more than 700 students from universities in eight countries, working in 80 teams on an EMIS software requirements task. Results indicate that students report positive learning experiences, with international students in international teams reporting the most positive experiences. There are also significant differences among universities.
|
Geoffrey Hoare, Jeffrey Nield, Tom Belcuore, & Tom Rich. (2008). Information needs and decision support in health and medical disasters. In B. V. de W. F. Fiedrich (Ed.), Proceedings of ISCRAM 2008 – 5th International Conference on Information Systems for Crisis Response and Management (pp. 778–786). Washington, DC: Information Systems for Crisis Response and Management, ISCRAM.
Abstract: During a disaster, health and medical decision makers need accurate, timely information. However, it is seldom readily available to the right decision makers, at the right time. Quite a number of databases currently exist with information about health and medical organizations which decision makers need during a disaster. Some of these databases have functions that facilitate decision-making and communication before, during and after a disaster. In theory, linking several existing databases will supply this information. Also, other functions can be provided in one package for incident management and monitoring of the preparedness capacity of a State's health and medical systems. But, this has not happened yet in Florida. This research assessed the different users needs, defined the information required to make good decisions and is testing a pilot decision support system of linked databases.
|
Marlen Hofmann, Stefan Sackmann, & Hans Betke. (2013). A novel architecture for disaster response workflow management systems. In J. Geldermann and T. Müller S. Fortier F. F. T. Comes (Ed.), ISCRAM 2013 Conference Proceedings – 10th International Conference on Information Systems for Crisis Response and Management (pp. 338–343). KIT; Baden-Baden: Karlsruher Institut fur Technologie.
Abstract: Due to the shared focus of disaster response management (DRM) and business process management on activities as well as the obvious similarity of disaster response processes (DRP) and business processes, the application of workflow management systems (WfMS) has been discussed as a promising approach to manage DRP. However, the application of WfMS in DRM has not yet been realized in practice. One reason for this is the lack of methods and tools in WfMS for taking interdependencies between activities, time, resource, and place into consideration. This considerably restricts the variety of DRP. Therefore, a novel architecture for a disaster response workflow management system is discussed. A special focus lies on the management and analysis of interdependencies that is seen as very promising to improve future DRM.
|
Katarína Zánická Holla, Ladislav Simak, & Jozef Ristvej. (2012). Systematic method of risk assessment in industrial processes. In Z.Franco J. R. L. Rothkrantz (Ed.), ISCRAM 2012 Conference Proceedings – 9th International Conference on Information Systems for Crisis Response and Management. Vancouver, BC: Simon Fraser University.
Abstract: The mankind has passed through a significant historical frontier, including technological processes where hazardous substances are the potential threats for the employees, the public, environment and property and therefore it is inevitable to pay increased attention to the occurrence of the industrial accidents and mainly to their prevention. The area of safety of technical and technological systems is very closely connected not only with the reliability of the processes but also with the technical as well as social, legal and other aspects. There exist several systematic approaches, methods and techniques to assess industrial processes and risks linked to these processes on the European Union level and also on the national level of each member state. We can mention some basic systematic approaches – MOSAR, CPQRA, ARAMIS, PRA (PSA) widely used in this area. But according to the analyst's needs they usually miss structured and systematic approach, how to complete the risk assessment in steps and what the content of each of these steps is. The main focus of this paper is to discuss the theoretical aspects of the risk assessment in industrial processes and to show the possible approach of structured and systematic methods with support of logical diagrams to fulfill all tasks concerning the industrial accident assessment. Last but not least it is to present the research activities of the Department of Crisis Management, University of Žilina in Žilina, Slovakia. © 2012 ISCRAM.
|
Hongmin Li, Doina Caragea, & Cornelia Caragea. (2017). Towards Practical Usage of a Domain Adaptation Algorithm in the Early Hours of a Disaster. In eds Aurélie Montarnal Matthieu Lauras Chihab Hanachi F. B. Tina Comes (Ed.), Proceedings of the 14th International Conference on Information Systems for Crisis Response And Management (pp. 692–704). Albi, France: Iscram.
Abstract: Many machine learning techniques have been proposed to reduce the information overload in social media data during an emergency situation. Among such techniques, domain adaptation approaches present greater potential as compared to supervised algorithms because they don't require labeled data from the current disaster for training. However, the use of domain adaptation approaches in practice is sporadic at best. One reason is that domain adaptation algorithms have parameters that need to be tuned using labeled data from the target disaster, which is presumably not available. To address this limitation, we perform a study on one domain adaptation approach with the goal of understanding how much source data is needed to obtain good performance in a practical situation, and what parameter values of the approach give overall good performance. The results of our study provide useful insights into the practical application of domain adaptation algorithms in real crisis situations.
|
Hongmin Li, Nicolais Guevara, Nic Herndon, Doina Caragea, Kishore Neppalli, Cornelia Caragea, et al. (2015). Twitter Mining for Disaster Response: A Domain Adaptation Approach. In L. Palen, M. Buscher, T. Comes, & A. Hughes (Eds.), ISCRAM 2015 Conference Proceedings ? 12th International Conference on Information Systems for Crisis Response and Management. Kristiansand, Norway: University of Agder (UiA).
Abstract: Microblogging data such as Twitter data contains valuable information that has the potential to help improve the speed, quality, and efficiency of disaster response. Machine learning can help with this by prioritizing the tweets with respect to various classification criteria. However, supervised learning algorithms require labeled data to learn accurate classifiers. Unfortunately, for a new disaster, labeled tweets are not easily available, while they are usually available for previous disasters. Furthermore, unlabeled tweets from the current disaster are accumulating fast. We study the usefulness of labeled data from a prior source disaster, together with unlabeled data from the current target disaster to learn domain adaptation classifiers for the target. Experimental results suggest that, for some tasks, source data itself can be useful for classifying target data. However, for tasks specific to a particular disaster, domain adaptation approaches that use target unlabeled data in addition to source labeled data are superior.
|
