Kejin Wang

TL;DR: This study develops a social media-disaster resilience framework, highlighting four contributions (communication, ground truth, sentiment analysis, predictive modeling) and four challenges (false information, disparities, data processing, algorithm bias), and proposes 20 strategies to improve social media use for disaster resilience.

TL;DR: The analysis of governmental Twitter accounts found that these accounts’ messages were most commonly meant to spread information about the pandemic, but that users were most likely to engage with tweets that requested readers take action, such as hand washing.

TL;DR: The authors analyzed Twitter activities during Hurricane Sandy in 2012, at both the county and the zip code area levels in the five affected states, and found that correlations between Twitter use indices and social-environmental variables representing community resilience still hold at the county level in previous studies, but they are weaker at the zip-code area level.

TL;DR: In this article , the authors evaluate a Random Forest regression model trained to predict Twitter rescue request rates from social-environmental data using three fairness criteria (independence, separation, and sufficiency).