Mining social media data to track environmental disaster events

Overview

Social media represents an emerging data resource for environmental disaster response and situational awareness. The increasing volume of social media usage during catastrophic events such as Hurricane Harvey (2017), Hurricane Ida (2021), and Hurricane Milton (2024) demonstrates potential value for disaster coordination and information dissemination. Extracting actionable intelligence from social media textual data presents both opportunities and challenges, including management of unstructured and ambiguous content, variability in user credibility, and data volume scaling.

Methods and approach

The research integrates three complementary analytical frameworks: textual classification of social media content through natural language processing and machine learning techniques, spatial and temporal analysis to situate events and information flow geographically and chronologically, and visual analytics to facilitate information presentation during rapid-response scenarios. This methodology aims to extract relevant information while filtering noise and misinformation from unstructured social media data.

Results

The abstract does not report empirical results or validation outcomes. Rather, it frames the research as a development initiative designed to address identified gaps in current disaster response capabilities. The proposed methodology remains at the stage of integration design, intended to enable rapid information processing and situational awareness during natural disaster events.

Implications

The synthesis of NLP, machine learning, spatial-temporal analysis, and visual analytics may provide infrastructure for more efficient information extraction during environmental disasters. Integration of these techniques could support coordination of relief efforts and enhance real-time situational awareness among response personnel. Development of such methodology could address documented challenges in managing high-volume, heterogeneous, and credibility-variable social media data streams.

Disclosure

• Research title: Mining social media data to track environmental disaster events
• Authors: Emiliano del Gobbo, Luigi Ippoliti, Lara Fontanella, Barbara Cafarelli
• Publication date: 2026-02-23
• DOI: https://doi.org/10.1007/s10651-025-00699-x
• OpenAlex record: View
• PDF: Download
• Image credit: Photo by Patrick Konior on Unsplash (Source • License)
• Disclosure: This post is an AI-generated summary of a research work. It was prepared by an editor. The original authors did not write or review this post.