Research Information System
IEEE TRANSACTIONS ON COMPUTATIONAL SOCIAL SYSTEMS, 2025 (SCI-Expanded, Scopus)
In recent years, social media has emerged as a primary channel for users to promptly share feedback and issues during disasters and emergencies, playing a key role in crisis management. While significant progress has been made in collecting and analyzing social media content, there remains a pressing need to enhance the automation, aggregation, and customization of this data to deliver actionable insights tailored to diverse stakeholders, including the press, police, EMS, firefighters, and other decision-makers. This effort is essential for improving the coordination of activities such as relief efforts, resource distribution, and media communication. This article presents a methodology that leverages the capabilities of large language models (LLMs) to enhance disaster response and management. Our approach combines classification techniques with generative AI to bridge the gap between raw user feedback and stakeholder-specific reports. Social media posts shared during catastrophic events, including earthquakes, floods, fires, and hurricanes, are analyzed with a focus on user-reported issues, service interruptions, and encountered challenges. We adopt a unified architecture that combines encoder-based models, such as BERT, for precise multidimensional classification, including content type, sentiment, emotion, geolocation, and topic, with decoder-based generative models, such as ChatGPT, to produce human-readable and informative reports tailored to different audiences. This integration of analytical and generative capabilities enables the synthesis of structured insights into coherent summaries. We compare standard approaches, which analyze posts directly using prompts in ChatGPT, to our advanced method, which incorporates multidimensional classification, sub-event selection, and tailored report generation. Our methodology demonstrates superior performance in both quantitative metrics, such as text coherence scores and latent representations, and qualitative assessments by automated tools and field experts, delivering precise, impactful insights for diverse disaster response stakeholders.