This research demonstrates how Large Language Models can transform social media data into customized, actionable insights during disasters and emergencies.

• Automates the analysis of social media content during crises to support real-time decision making
• Delivers tailored information to different emergency stakeholders (police, firefighters, EMS, press)
• Creates a unified framework that extracts, categorizes, and prioritizes critical information from unstructured social media posts
• Enhances security response capabilities by providing timely, relevant insights to those managing disaster situations

This innovation bridges the gap between abundant social media data and practical emergency response needs, potentially saving lives through more informed and coordinated crisis management.

Multi-Stakeholder Disaster Insights from Social Media Using Large Language Models