Social media for enhanced understanding of disaster resilience during Hurricane Florence

Faxi Yuan*, Min Li, Rui Liu*, Wei Zhai, Bing Qi

*Corresponding author for this work

    Research output: Contribution to journalJournal articlepeer-review

    66 Citations (Scopus)

    Abstract

    Citizens with different demographic characters presented varying responses and behaviors in the same disasters. Their divergent responses can impact their actual damages during crises. Previous studies have employed social media for analyzing citizens’ crisis responses. However, these studies missed the demographic dimension. To resolve this limitation, this research proposes three objectives: 1) to explore the variances of sentiment polarities among different racial/ethnic and gender groups; 2) to investigate the concern themes in their expressions, including theme popularity and their sentiment towards these themes; 3) to enhance the understanding of social aspects of disaster resilience with the results of disaster response disparities. Results indicate that Hispanic and male groups are more likely to express negative sentiment. The black group pays the least attention to ‘hurricane warn’ and shows most interests in ‘pray/donate’. The white group is most optimistic about hurricane/flood impacts while the black group shows dissatisfaction towards ‘response’. The female group pays less attention to ‘hurricane warn’ while they are more optimistic towards ‘hurricane/flood impact’ and ‘response’ than the male group. Our findings can help crisis response managers identify the more sensitive/vulnerable groups in the crisis and provide on-target disaster evolution reports and relief resources to the corresponding demographic groups.
    Original languageEnglish
    Article number102289
    Number of pages18
    JournalInternational Journal of Information Management
    Volume57
    Early online date22 Dec 2020
    DOIs
    Publication statusPublished - Apr 2021

    User-Defined Keywords

    • Disaster resilience
    • Social media
    • Sentiment analysis
    • Data mining
    • LDA topic model
    • Situation awareness

    Fingerprint

    Dive into the research topics of 'Social media for enhanced understanding of disaster resilience during Hurricane Florence'. Together they form a unique fingerprint.

    Cite this