Scholarlink

<p>The increasing prevalence of social media usage has led to the emergence of mining social media data as a valuable resource for disaster response. Mining their textual data presents opportunities and challenges. Advanced techniques in natural language processing (NLP) and machine learning enable the extraction of relevant information while effectively filtering out noise and misinformation. Real-world cases, such as Hurricane Harvey (2017), Hurricane Ida (2021), Hurricane Milton (2024), and Hurricane Melissa (2025) highlight the important role of social media in coordinating disaster relief efforts and enhancing situational awareness. Challenges include unstructured and ambiguous data, diverse user credibility, and overwhelming data volume. The aim of this research is to develop a methodology that integrates textual classification of social media data, spatial and temporal analysis, and visual analytics to provide rapid responses during natural disasters.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Mining social media data to track environmental disaster events

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年2月23日

查看全文
  • Emiliano del Gobbo,
  • Luigi Ippoliti,
  • Lara Fontanella,
  • Barbara Cafarelli

摘要

The increasing prevalence of social media usage has led to the emergence of mining social media data as a valuable resource for disaster response. Mining their textual data presents opportunities and challenges. Advanced techniques in natural language processing (NLP) and machine learning enable the extraction of relevant information while effectively filtering out noise and misinformation. Real-world cases, such as Hurricane Harvey (2017), Hurricane Ida (2021), Hurricane Milton (2024), and Hurricane Melissa (2025) highlight the important role of social media in coordinating disaster relief efforts and enhancing situational awareness. Challenges include unstructured and ambiguous data, diverse user credibility, and overwhelming data volume. The aim of this research is to develop a methodology that integrates textual classification of social media data, spatial and temporal analysis, and visual analytics to provide rapid responses during natural disasters.