Gayane Shalunts, Gerhard Backfried, & Prinz Prinz. (2014). Sentiment analysis of German social media data for natural disasters. 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. 752–756). University Park, PA: The Pennsylvania State University.
Abstract: Analysis of social media and traditional media provides significant information to first responders in times of natural disasters. Sentiment analysis, particularly of social media originating from the affected population, forms an integral part of multifaceted media analysis. The current paper extends an existing methodology to the domain of natural disasters, broadens the support of multiple languages and introduces a new manner of classification. The performance of the approach is evaluated on a recently collected dataset manually annotated by three human annotators as a reference. The experiments show a high agreement rate between the approach taken and the annotators. Furthermore, the paper presents the initial application of the resulting technology and models to sentiment analysis of social media data in German, covering data collected during the Central European floods of 2013.
|
Spyros Chrysanthopoulos, Theofanis Kapetanakis, Giannis Chaidemenos, Stelios Vernardos, Harris Georgiou, & Claudio Rossi. (2020). Emergency Response in Recent Urban/Suburban Disaster Events in Attica: Technology Gaps, Limitations and Lessons Learned. 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. 984–989). Blacksburg, VA (USA): Virginia Tech.
Abstract: Emergency response operations in large-scale urban/suburban disaster events is often addressed by the standard protocols and international guidelines for collapsed buildings, heavy debris, etc. However, a wide range of First Responder (FR) operations need to address various other contexts, work environments and hazards. In this paper, two real disaster events are explored as use cases for such urban/suburban FR operations, namely a flash flood and a wildfire, both in Attica, Greece (2017-2018). Based on our team's experience from these mobilizations and active participation in both these events as FR actor in the field, we present the challenges, the complexity of such multi-aspect disaster events, the limitations of emergency response, the technology gaps of the FR teams, as well as the lessons learned during these deployments. Finally, we make some notes on future prospects and possible advancements in tools and technologies that would greatly enhance the operational safety and readiness of the FR teams in such events.
|
Milica Stojmenovic, & Gitte Lindgaard. (2014). Probing PROBE: A field study of an advanced decision support prototype for managing chemical, biological, radiological, nuclear, and explosives (CBRNE) events. 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. 90–99). University Park, PA: The Pennsylvania State University.
Abstract: The purpose of this field study was investigate teamwork and communication among event management personnel, to assess the degree to which PROBE, the advanced prototype they were using to manage a CBRNE simulation, would adequately meet their needs. The study was a continuation of previous research conducted in the early phase of PROBE development. Two communication-related analyses were applied to identify instances of effective and of ineffective communication among the management team. These revealed that communication was mostly effective. However, the one serious communication breakdown that was observed could have had fatal consequences. It showed that great care must be taken to ensure the safety of first responders at all times when evaluating prototypes in the field. A list of questions was generated from the lessons learned to assist future researchers prepare for CBRNE field studies.
|
Andrea H. Tapia, Nicolas LaLone, & Hyun-Woo Kim. (2014). Run amok: Group crowd participation in identifying the bomb and bomber from the Boston marathon bombing. 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. 265–274). University Park, PA: The Pennsylvania State University.
Abstract: In this paper we tell a version of the story of the bombing of the Boston Marathon. At first, two online groups gathered images, video and textual information concerning the bombing of the Boston Marathon and shared these with the FBI and amongst themselves. Secondly, these groups then created mechanisms to conduct their own investigation into the identities of the perpetrators. Finally, the larger national media followed the results of these online group investigations and reported these as fact to a national audience. We choose Twitter as our data repository and conducted quantitative analyses of tweets sent during the Boston Bombing. The implications for not incorporating public crowd participation within the standard operating procedures of emergency services may result in either a loss of public confidence in the slow-moving nature of official response to uncontrollable, dangerous and irresponsible public and media participation that exacerbates the negative effects of any disaster.
|
Thomas Theodoridis, George Katsikas, Nicholas Vretos, & Petros Daras. (2022). A Symbiotic Orchestration Module for Multi-agent Collaboration in Disaster Response Scenarios. In Rob Grace, & Hossein Baharmand (Eds.), ISCRAM 2022 Conference Proceedings – 19th International Conference on Information Systems for Crisis Response and Management (pp. 771–780). Tarbes, France.
Abstract: This paper presents the Symbiotic Orchestration Module, which facilitates the collaboration of smart agents in disaster response scenarios. By effectively orchestrating the actions of different agents in critical situations towards a common goal, it enhances the individual capabilities of the agents and unlocks new possibilities that are not available when agents act isolated. To achieve this, the Symbiotic Orchestration Module is composed of four sub-modules: a) the Mission Controller, which is responsible for keeping track of ongoing missions, agent allocations and for handling non-collaborative missions, b) the Symbiotic Operation Control Module, which handles collaborative missions proposed by the system, c) the Task Allocation Module, which automatically assigns available robots to incoming missions based on robot capabilities and mission requirements, and d) the Task Recognition and Optimal Sequencing Module, which is responsible for recognizing opportunities for agent collaboration and for system-wide goal optimization.
|
Tolt, G., Rydell, J., Bilock, E., Eek, J., Andersson, P., & Nygårds, J. (2023). Real-time Multi-Sensor Positioning for First Responders. In Jaziar Radianti, Ioannis Dokas, Nicolas Lalone, & Deepak Khazanchi (Eds.), Proceedings of the 20th International ISCRAM Conference (pp. 177–187). Omaha, USA: University of Nebraska at Omaha.
Abstract: This paper describes a concept for real-time positioning of first responders that includes a number of complementary sensors worn by the first responder, to increase accuracy and robustness in indoor and complex environments. By using sensors of different types, each with their own strengths and limitations, and fusing their respective outputs, the goal is to increase the usability of positioning information in time-critical and risky operations. This facilitates synchronization of activities and increases safety in the operation. The sensors included in the proposed real-time positioning module are shoe-mounted inertial measurement units, ultra-wideband radio, thermal and visual cameras, and GNSS. The fusion framework is based on factor graphs. This work-in-progress paper describes the individual sensor components and shows preliminary findings concerning the possibilities to improve position estimation through sensor fusion.
|
Yulia Tyshchuk, & William A. Wallace. (2013). The use of social media by local government in response to an extreme event: Del norte county, CA response to the 2011 Japan tsunami. 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. 802–811). KIT; Baden-Baden: Karlsruher Institut fur Technologie.
Abstract: Social media has become increasingly important for emergency management. One example is its current use by governmental organizations to disseminate emergency-relevant information. During disaster events, it is imperative for people in affected areas to obtain accurate information. People using social media make a conscious decision to trust, act on, propagate or disregard emergency-relevant information. However, local government, in general, has not developed agreed upon ways to use social media in emergencies. This study documents how emergency management was able to successfully partner with local media and utilize social media to develop important relationships with the affected community via social media in emergencies. The study demonstrates a way to successfully utilize social media during disaster events in several ways: by closing a feedback loop between first responders and the public, by monitoring information flow, and by providing regular updates to the public.
|