Li, H., Caragea, D., Mhatre, A., Ge, J., & Liu, M. (2023). Identifying COVID-19 Tweets Relevant to Low-Income Households Using Semi-supervised BERT and Zero-shot ChatGPT Models. In Jaziar Radianti, Ioannis Dokas, Nicolas Lalone, & Deepak Khazanchi (Eds.), Proceedings of the 20th International ISCRAM Conference (pp. 953–963). Omaha, USA: University of Nebraska at Omaha.
Abstract: Understanding the COVID-19 pandemic impacts on low-income households can inform social services about the needs of vulnerable communities. Some recent works have studied such impacts through social media content analysis, and supervised machine learning models have been proposed to automatically classify COVID-19 tweets into different categories, such as income and economy impacts, social inequality and justice issues, etc. In this paper, we propose semi-supervised learning models based on BERT with Self-Training and Knowledge Distillation for identifying COVID-19 tweets relevant to low-income households by leveraging readily available unlabeled data in addition to limited amounts of labeled data. Furthermore, we explore ChatGPT’s potential for annotating COVID-19 data and the performance of fine-tuned GPT-3 models. Our semi-supervised BERT model with Knowledge Distillation showed improvements compared to a supervised baseline model, while zero-shot ChatGPT showed good potential as a tool for annotating crisis data. However, our study suggests that the cost of fine-tuning large and expensive GPT-3 models may not be worth for some tasks.
|
|
Berggren, P., Ryrberg, T., Lindhagen, A., & Johansson, B. (2023). Building capacity – conceptualizing Training of Trainers. In Jaziar Radianti, Ioannis Dokas, Nicolas Lalone, & Deepak Khazanchi (Eds.), Proceedings of the 20th International ISCRAM Conference (pp. 701–710). Omaha, USA: University of Nebraska at Omaha.
Abstract: Many organizations train and educate their staff to prepare for crisis. One approach is train-the-trainer (ToT; Training of trainers) concept. It is based on the idea that someone can be trained as a trainer, who in turn train their colleagues. The philosophy resembles a pyramid scheme that allows for a fast and efficient spread of knowledge and skills. This study focused on perceptions of the ToT concept through interviews with ToT trainers. Two learning theories, organizational learning (4I) and experiential learning theory (ELT) were used to conceptualize the ToT-concept. It was found that the ToT-concept can be used as the method to conduct ELT to achieve organizational learning and knowledge (4I). Furthermore, the study also presents how participants perceives ToT using thematic analysis. This resulted in four themes: Common understanding of ToT, Learn-by doing, No grounding in ToT, and Difficult to ensure quality.
|
|
Steen-Tveit, K., Snaprud, M. H., Heinecke, J. E., & Fure Nora. (2023). Towards a Co-Created Emergency Management Collaboration Repository. In Jaziar Radianti, Ioannis Dokas, Nicolas Lalone, & Deepak Khazanchi (Eds.), Proceedings of the 20th International ISCRAM Conference (pp. 20–32). Omaha, USA: University of Nebraska at Omaha.
Abstract: The need for information systems (ISs) to aid emergency management (EM) has been well established. Yet, despite the acknowledged benefits of ISs for EM, the support of ISs in the preparedness phase is weak. Complex EM operations require coordinated efforts across emergency organizations, which are facing enormous challenges related to the method of collaboration to cope with the impact. This paper presents an ongoing project initiated to develop an emergency management collaboration repository for a range of emergency responders, focusing on emergency cross-organizational collaboration, information sharing, exercises, and evaluations. A participatory design approach was applied for the system requirements elicitation and was carried out in two workshops with several EM stakeholders.
|
|
Peng Xia, Ji Ruan, Dave Parry, Jian Yu, & Sally Britnell. (2023). Enhancing Triage Training for Mass Casualty Incidents with Virtual Reality and Artificial Intelligence. In V. L. Thomas J. Huggins (Ed.), Proceedings of the ISCRAM Asia Pacific Conference 2022 (pp. 68–76). Palmerston North, New Zealand: Massey Unversity.
Abstract: Mass casualty incidents (MCIs) occur with natural or man-made disasters. Training emergency staff for combating MCIs is essential, but the cost can be high as such incidents rarely occur, and a physical simulation is resource-intensive. Triage is a critical task in dealing with MCIs. In this paper, we propose to use Virtual Reality (VR) and Artificial Intelligence (AI) technologies to build a low-cost, high-efficient system for MCI triage training. Our system captures more comprehensive training data and utilizes state-of-the-art AI evaluation methods.
|
|
Pauline Tobergte, Alena Knispel, Lennart Landsberg, & Ompe Aimé Mudimu. (2022). Evaluation of Tabletop Exercises in Emergency Response Research and Application in the Research Project SORTIE. In Rob Grace, & Hossein Baharmand (Eds.), ISCRAM 2022 Conference Proceedings – 19th International Conference on Information Systems for Crisis Response and Management (pp. 415–427). Tarbes, France.
Abstract: This paper presents the fields of application of the tabletop exercise in emergency response by explaining the method in emergency response research. The authors illustrate the tabletop exercise of the Institute for Rescue Engineering and Civil Protection (IRG) of the TH Köln in a research project on Sensor Systems for Localization of Trapped Victims in Collapsed Infrastructure (acronym: SORTIE) as an application example. Subsequently, the quantitative and qualitative evaluation methods used generally and specifically for the tabletop exercise of the research project SORTIE are considered, and the technical implementation is explained. The evaluation method used in the tabletop exercise consists of three sub-areas (participant survey; exercise observation; photo, video and audio recordings). Further, the analysis of the evaluation using statistical tools is explained. Finally, this paper refers to possible sources of error in the evaluation of tabletop exercises, such as exercise artificiality and subjectivity of the exercise observers.
|
|