Search

Social media offers a unique lens to observe large-scale, spatial-temporal patterns of users' reactions toward critical events. However, social media use varies across demographics, with younger users being more prevalent compared to older populations. This difference introduces biases in data representativeness, and analysis based on social media without proper adjustment will lead to overlooking the voices of digitally marginalized communities and inaccurate estimations. This study explores solutions to pinpoint and alleviate the demographic biases in social media analysis through a case study estimating the public sentiment about COVID-19 using Twitter data. We analyzed the pandemic-related Twitter data in the U.S. during 2020–2021 to (1) elucidate the uneven social media usage among demographic groups and the disparities of their sentiments toward COVID-19, (2) construct an adjusted public sentiment measurement based on social media, the Sentiment Adjusted by Demographics (SAD) index, to evaluate the spatiotemporal varying public sentiment toward COVID-19. The results show higher proportions of female and adolescent Twitter users expressing negative emotions to COVID-19. The SAD index unveils that the public sentiment toward COVID-19 was most negative in January and February 2020 and most positive in April 2020. Vermont and Wyoming were the most positive and negative states toward COVID-19. [ABSTRACT FROM AUTHOR]

Extracting precise geographical information from the textual content, referred to as toponym recognition, is fundamental in geographical information retrieval and crucial in a plethora of spatial analyses, e.g. mining location-based information from social media, news reports, and surveys for various applications. However, the performance of existing toponym recognition methods and tools is deficient in supporting tasks that rely on extracting fine-grained geographic information from texts, e.g. locating people sending help requests with addresses through social media during disasters. The emerging pretrained language models have revolutionized natural language processing and understanding by machines, offering a promising pathway to optimize toponym recognition to underpin practical applications. In this paper, TopoBERT, a uniquely designed toponym recognition module based on a one-dimensional Convolutional Neural Network (CNN1D) and Bidirectional Encoder Representation from Transformers (BERT), is proposed and fine-tuned. Three datasets are leveraged to tune the hyperparameters and discover the best strategy to train the model. Another seven datasets are used to evaluate the performance. TopoBERT achieves state-of-the-art performance (average f1-score = 0.854) compared to the seven baseline models. It is encapsulated into easy-to-use python scripts and can be seamlessly applied to diverse toponym recognition tasks without additional training. [ABSTRACT FROM AUTHOR]

This paper develops a social media-disaster resilience analysis framework by categorizing types of social media use and their challenges to better understand and assess its role in disaster resilience research and management. The framework is derived primarily from several case studies of Twitter use in three hurricane events in the United States – Hurricanes Isaac, Sandy, and Harvey. The paper first outlines four major contributions of social media data for disaster resilience research and management, which include serving as an effective communication platform, providing ground truth information for emergency response and rescue operations, providing information on people's sentiments, and allowing predictive modeling. However, there are four key challenges to its uses, which include easy spreading of false information, social and geographical disparities of Twitter use, technical issues on processing and analyzing big and noisy data, especially on improving the locational accuracy of the tweets, and algorithm bias in AI and other types of modeling. Then, the paper proposes twenty strategies that the four sectors of the social media community – organizations, individuals, social media companies, and researchers – could take to improve social media use to increase disaster resilience. [ABSTRACT FROM AUTHOR]

The Covid-19 has presented an unprecedented challenge to public health worldwide. However, residents in different countries showed diverse levels of Covid-19 awareness during the outbreak and suffered from uneven health impacts. This study analyzed the global Twitter data from January 1st to June 30th, 2020, to answer two research questions. What are the linguistic and geographical disparities of public awareness in the Covid-19 outbreak period reflected on social media? Does significant association exist between the changing Covid-19 awareness and the pandemic outbreak? We established a Twitter data mining framework calculating the Ratio index to quantify and track awareness. The lag correlations between awareness and health impacts were examined at global and country levels. Results show that users presenting the highest Covid-19 awareness were mainly those tweeting in the official languages of India and Bangladesh. Asian countries showed more disparities in awareness than European countries, and awareness in Eastern Europe was higher than in central Europe. Finally, the Ratio index had high correlations with global mortality rate, global case fatality ratio, and country-level mortality rate, with 21-31, 35-42, and 13–18 leading days, respectively. This study yields timely insights into social media use in understanding human behaviors for public health research. [ABSTRACT FROM AUTHOR]

Despite the increasingly prominent role of social media in disaster events, studies analyzing its use in rescue operations remain scanty. Hurricane Harvey hit Texas with unprecedented rainfall and flooding in 2017 and was marked by widespread use of social media for rescue requests. We conducted a survey of 195 Twitter users in Houston and surrounding communities who had requested for rescue during Harvey. The objective was to investigate our targeted group's socioeconomic and flood exposure characteristics, report the effectiveness of Twitter, and highlight lessons learnt and suggestions made for its use in future rescue missions. Survey revealed that those requesting rescue on Twitter were better educated, employed (80%), and homeowners (81%). Majority of them were flooded (87%), but remained satisfied with current location and did not consider moving. Calling relatives and friends for rescue was most responsive and yielded higher assistance-provided rate than using Twitter. Our respondents found Twitter helpful, but identified issues such as not knowing when volunteers received their requests or whether they would send help. They suggested promoting Twitter accounts and hashtags that accept emergency requests. This study provides baseline information and actionable suggestions for first responders, community managers, and resilience practitioners to improve future rescue missions. [ABSTRACT FROM AUTHOR]

Social media such as Twitter is increasingly being used as an effective platform to observe human behaviors in disastrous events. However, uneven social media use among different groups of population in different regions could lead to biased consequences and affect disaster resilience. This paper studies the Twitter use during 2017 Hurricane Harvey in 76 counties in Texas and Louisiana. We seek to answer a fundamental question: did social-geographical disparities of Twitter use exist during the three phases of emergency management (preparedness, response, recovery)? We employed a Twitter data mining framework to process the data and calculate two indexes: Ratio and Sentiment. Regression analyses between the Ratio indexes and the social-geographical characteristics of the counties at the three phrases reveal significant social and geographical disparities in Twitter use during Hurricane Harvey. Communities with higher disaster-related Twitter use in Harvey generally were communities having better social and geographical conditions. These results of Twitter use patterns can be used to compare with future similar studies to see whether the Twitter use disparities have increased or decreased. Future research is also needed to examine the effects of Twitter use disparities on disaster resilience and to test whether Twitter use can predict community resilience. [ABSTRACT FROM AUTHOR]

Coastal communities faced with multiple hazards have shown uneven responses and behaviors. These responses and behaviors could be better understood by analyzing real-time social media data through categorizing them into the three phases of the emergency management: preparedness, response, and recovery. This study analyzes the spatial-temporal patterns of Twitter activities during Hurricane Sandy, which struck the U.S. Northeast on 29 October 2012. The study area includes 126 counties affected by Hurricane Sandy. The objectives are threefold: (1) to derive a set of common indexes from Twitter data so that they can be used for emergency management and resilience analysis; (2) to examine whether there are significant geographical and social disparities in disaster-related Twitter use; and (3) to test whether Twitter data can improve postdisaster damage estimation. Three corresponding hypotheses were tested. Results show that common indexes derived from Twitter data, including ratio, normalized ratio, and sentiment, could enable comparison across regions and events and should be documented. Social and geographical disparities in Twitter use existed in the Hurricane Sandy event, with higher disaster-related Twitter use communities generally being communities of higher socioeconomic status. Finally, adding Twitter indexes into a damage estimation model improved the adjusted R2 from 0.46 to 0.56, indicating that social media data could help improve postdisaster damage estimation, but other environmental and socioeconomic variables influencing the capacity to reducing damage might need to be included. The knowledge gained from this study could provide valuable insights into strategies for utilizing social media data to increase resilience to disasters. [ABSTRACT FROM AUTHOR]

Social media plays increasingly significant roles in disaster response, but effectively leveraging social media for rescue is challenging. This study analyzed rescue requests on Twitter during the 2017 Hurricane Harvey, in which many residents resorted to social media to call for help. The objectives include: (1) understand the characteristics of rescue-request messages; (2) reveal the spatial-temporal patterns of rescue requests; (3) determine the social-geographical conditions of communities needing rescue; and (4) identify the challenges of using social media for rescue and propose improvement strategies. About half of rescue requests either did not provide sufficient information or neglected to include rescue-related hashtags or accounts. Of the 824 geocoded unique rescue requests, 41% were from FEMA-defined minimal flood risk zones. Communities sending more rescue requests on Twitter were environmentally and socioeconomically more vulnerable. Finally, we derived a framework summarizing the steps and strategies needed to improve social media use for rescue operations. [Display omitted] • Explained who, how, and why social media were used as a platform for emergency rescue. • Revealed the spatial-temporal patterns of rescue requests during 2017 Hurricane Harvey. • Determine the underlying geographical and social conditions of communities needing rescue. • Identified the challenges of using social media for rescue and proposed improvement suggestions. [ABSTRACT FROM AUTHOR]

Social media platforms are playing increasingly critical roles in disaster response and rescue operations. During emergencies, users can post rescue requests along with their addresses on social media, while volunteers can search for those messages and send help. However, efficiently leveraging social media in rescue operations remains challenging because of the lack of tools to identify rescue request messages on social media automatically and rapidly. Analyzing social media data, such as Twitter data, relies heavily on Natural Language Processing (NLP) algorithms to extract information from texts. The introduction of bidirectional transformers models, such as the Bidirectional Encoder Representations from Transformers (BERT) model, has significantly outperformed previous NLP models in numerous text analysis tasks, providing new opportunities to precisely understand and classify social media data for diverse applications. This study developed and compared ten VictimFinder models for identifying rescue request tweets, three based on milestone NLP algorithms and seven BERT-based. A total of 3191 manually labeled disaster-related tweets posted during 2017 Hurricane Harvey were used as the training and testing datasets. We evaluated the performance of each model by classification accuracy, computation cost, and model stability. Experiment results show that all BERT-based models have significantly increased the accuracy of categorizing rescue-related tweets. The best model for identifying rescue request tweets is a customized BERT-based model with a Convolutional Neural Network (CNN) classifier. Its F1-score is 0.919, which outperforms the baseline model by 10.6%. The developed models can promote social media use for rescue operations in future disaster events. • Manually labeled a dataset that trains NLP models for classifying rescue requesting tweets. • Evaluated the performance of 10 different models, including three based on milestone NLP algorithms and seven BERT-based models. • Designed a customized CNN classifier placed on the top of BERT model that achieved best performance. • Laid the foundation of developing future victim finding applications. [ABSTRACT FROM AUTHOR]

Disaster resilience is the capacity of a community to "bounce back" from disastrous events. Most studies rely on traditional data such as census data to study community resilience. With increasing use of social media, new data sources such as Twitter could be utilized to monitor human response during different phases of disasters to better understand resilience. An important research question is: Does Twitter use correlate with disaster resilience? Specifically, will communities with more disaster-related Twitter uses be more resilient to disasters, presumably because they have better situational awareness? The underlying issue is that if there are social and geographical disparities in Twitter use, how will such disparities affect communities' resilience to disasters? This study examines the relationship between Twitter use and community resilience during Hurricane Isaac, which hit Louisiana and Mississippi in August 2012. First, we applied the resilience inference measurement (RIM) model to calculate the resilience indices of 146 affected counties. Second, we analyzed Twitter use and their sentiment patterns through the three phases of Hurricane Isaac—preparedness, response, and recovery. Third, we correlated Twitter use density and sentiment scores with the resilience scores and major social–environmental variables to test whether significant geographical and social disparities in Twitter use existed through the three phases of disaster management. Significant positive correlations were found between Twitter use density and resilience indicators, confirming that communities with higher resilience capacity, which are characterized by better social–environmental conditions, tend to have higher Twitter use. These results imply that Twitter use during disasters could be improved to increase the resilience of affected communities. On the other hand, no significant correlations were found between sentiment scores and resilience indicators, suggesting that further research on sentiment analysis may be needed. [ABSTRACT FROM AUTHOR]