The increasing popularity of multimedia messages shared through public or private social media spills into diverse information dissemination contexts. To date, public social media has been explored as a potential alert system during natural disasters, but high levels of noise (i.e., non-relevant content) present challenges in both understanding social experiences of a disaster and in facilitating disaster recovery. This study builds on current research by uniquely using social media data, collected in the field through qualitative interviews, to create a supervised machine learning model. Collected data represents rescuers and rescuees during the 2017 Hurricane Harvey. Preliminary findings indicate a 99% accuracy in classifying data between signal and noise for signal-to-noise ratios (SNR) of 1:1, 1:2, 1:4, and 1:8. We also find 99% accuracy in classification between respondent types (volunteer rescuer, official rescuer, and rescuee). We furthermore compare human and machine coded attributes, finding that Google Vision API is a more reliable source of detecting attributes for the training set.
more »
« less
Evaluating the performance of Deep learning methods for hurricane-related image classification.
Global social media use during natural disasters has been well documented (Murthy et al., 2017). In the U.S., public social media platforms are often a primary venue for those affected by disasters . Some disaster victims believe first responders will see their public posts and that the 9-1-1 telephone system becomes overloaded during crises. Moreover, some feel that the accuracy and utility of information on social media is likely higher than traditional media sources . However, sifting through content during a disaster is often difficult due to the high volume of ‘non-relevant’ content. In addition, text is studied more than images posted on Twitter, leaving a potential gap in understanding disaster experiences. Images posted on social media during disasters have a high level of complexity (Murthy et al., 2016). Our study responds to O’Neal et al.’s (2017) call-to-action that social media images posted during disasters should be studied using machine learning.
more »
« less
- Award ID(s):
- 1760453
- NSF-PAR ID:
- 10120493
- Date Published:
- Journal Name:
- Proceedings of the ... International ISCRAM Conference
- ISSN:
- 2411-3387
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
more » « less
Abstract Researchers have established the prominent role digital volunteers play during crises and disasters. From self‐organizing to annotating public data, these volunteers are now a fixture in disaster research. However, we know much less about how these volunteers function, behind the public scene, when using private social media as a disaster unfolds and people need to be rescued. This qualitative study identified the emergent helping roles along with the skillsets and abilities that helped volunteers perform these behind‐the‐scene roles during the Hurricane Harvey flooding in 2017. Using in‐depth interviews along with captured images in private social media, we find these volunteers resembled organizational knowledge workers. We identify nine specific communicating and coordinating actions that these disaster knowledge workers performed. The contributions of these findings centre on implications for disaster response and management.
-
Social media is being increasingly utilized to spread breaking news and updates during disasters of all magnitudes. Unfortunately, due to the unmoderated nature of social media platforms such as Twitter, rumors and misinformation are able to propagate widely. Given this, a surfeit of research has studied rumor diffusion on social media, especially during natural disasters. In many studies, researchers manually code social media data to further analyze the patterns and diffusion dynamics of users and misinformation. This method requires many human hours, and is prone to significant incorrect classifications if the work is not checked over by another individual. In our studies, we fill the research gap by applying seven different machine learning algorithms to automatically classify misinformed Twitter data that is spread during disaster events. Due to the unbalanced nature of the data, three different balancing algorithms are also applied and compared. We collect and drive the classifiers with data from the Manchester Arena bombing (2017), Hurricane Harvey (2017), the Hawaiian incoming missile alert (2018), and the East Coast US tsunami alert (2018). Over 20,000 tweets are classified based on the veracity of their content as either true, false, or neutral, with overall accuracies exceeding 89%.more » « less
-
Lankes, R.David (Ed.)Resilience is often treated as a single-dimension system attribute, or various dimensions of resilience are studied separately without considering multi-dimensionality. The increasing frequency of catastrophic natural or man-made disasters affecting rural areas demands holistic assessments of community vulnerability and assessment. Disproportionate effects of disasters on minorities, low-income, hard-to-reach, and vulnerable populations demand a community-oriented planning approach to address the “resilience divide.” Rural areas have many advantages, but low population density, coupled with dispersed infrastructures and community support networks, make these areas more affected by natural disasters. This paper will catalyze three key learnings from our current work in public librarians’ roles in disaster resiliency: 1) rural communities are composed of diverse sub-communities, each which experiences and responds to traumatic events differently, depending on micro-geographic and demographic drivers; 2) public libraries are central to rural life, providing a range of informational, educational, social, and personal services, especially in remote areas that lack reliable access to community resources during disasters; and 3) rural citizens tend to be very self-reliant and are committed to strengthening and sustaining community resiliency with local human capital and resources. Public libraries and their librarian leaders are often a “crown jewel” of rural areas’ community infrastructure and this paper will present a community-based design and assessment process for resiliency hubs located in and operated through rural public libraries. The core technical and social science research questions explored in the proposed paper are: 1) Who were the key beneficiaries and what did they need? 2) What was the process of designing a resiliency hub? 3) What did library resiliency hubs provide and how can they be sustained? This resiliency hub study will detail co-production of solutions and involves an inclusive collaboration among researchers, librarians, and community members to address the effects of cascading impacts of natural disasters. The novel co-design process detailed in the paper reflects 1) an in-depth understanding of the complex interactions among libraries, residents, governments, and other agencies by collecting sociotechnical hurricane-related data for Calhoun County, Florida, USA, a region devastated by Hurricane Michael (2018) and hard-hit by Covid-19; 2) analyzed data from newly-developed fusing algorithms and incorporating multiple communities; and 3) co-designed resiliency hubs sited in public libraries. This research leverages a unique opportunity for the co-development of integrated library-centered policies and technologies to establish a new paradigm for developing disaster resiliency in rural settings. Public libraries serve a diverse population who will directly benefit from practical support tailored to their needs. The project will inform efficient plans to ensure that high-need groups are not isolated in disasters. The knowledge and insight gained from disseminating the study’s results will not only improve our understanding of emergency response operations, but also will contribute to the development of new disaster-related policies and plans for public libraries, with a broader application to rural communities in many settings.more » « less
-
Lankes, R. David (Ed.)Resilience is often treated as a single-dimension system attribute, or various dimensions of resilience are studied separately without considering multi-dimensionality. The increasing frequency of catastrophic natural or man-made disasters affecting rural areas demands holistic assessments of community vulnerability and assessment. Disproportionate effects of disasters on minorities, low-income, hard-to-reach, and vulnerable populations demand a community-oriented planning approach to address the “resilience divide.” Rural areas have many advantages, but low population density, coupled with dispersed infrastructures and community support networks, make these areas more affected by natural disasters. This paper will catalyze three key learnings from our current work in public librarians’ roles in disaster resiliency: rural communities are composed of diverse sub-communities, each which experiences and responds to traumatic events differently, depending on micro-geographic and demographic drivers. Rural citizens tend to be very self-reliant and are committed to strengthening and sustaining community resiliency with local human capital and resources. Public libraries are central to rural life, providing a range of informational, educational, social, and personal services, especially in remote areas that lack reliable access to community resources during disasters. Public libraries and their librarian leaders are often a “crown jewel” of rural areas’ community infrastructure and this paper will present a community-based design and assessment process for resiliency hubs located in and operated through rural public libraries. The core technical and social science research questions explored in the proposed paper are: 1) Who were the key beneficiaries and what did they need? 2) What was the process of designing a resiliency hub? 3) What did library resiliency hubs provide and how can they be sustained? This resiliency hub study will detail co-production of solutions and involves an inclusive collaboration among researchers, librarians, and community members to address the effects of cascading impacts of natural disasters. The novel co-design process detailed in the paper reflects an in-depth understanding of the complex interactions among libraries, residents, governments, and other agencies by collecting sociotechnical hurricane-related data for Calhoun County, Florida, USA, a region devastated by Hurricane Michael (2018) and hard-hit by Covid-19. We analyzed data from newly developed fusing algorithms and incorporating multiple communities and developed a framework and process to co-design resiliency hubs sited in public libraries. This research leverages a unique opportunity to library-centered policies and technologies to establish a new paradigm for developing disaster resiliency in rural settings. Public libraries serve a diverse population who will directly benefit from practical support tailored to their needs. The project will inform efficient plans to ensure that high-need groups are not isolated in disasters. The knowledge and insight gained from the resiliency hub design process will not only improve our understanding of emergency response operations, but also will contribute to the development of new disaster related policies and plans for public libraries, with a broader application to rural communities in many settings.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Conference Paper:
- The DOI is not currently available.
