Large Language Models for Causal Relations Extraction in Social Media: A Validation Framework for Disaster Intelligence

2026-05-11

AI Summary

This study addresses the challenge of extracting causal relationships from unstructured text in social media disaster-related posts by proposing the first causality verification framework tailored for disaster intelligence. The framework integrates large language models with expert knowledge, constructing causal graphs and aligning them with authoritative post-disaster reports to distinguish model hallucinations from genuine causal signals. Experimental results demonstrate that the approach effectively evaluates the evidential support for extracted causal claims, thereby revealing both the potential and the risks of large language models in disaster situational awareness. By grounding causal inferences in verifiable evidence, this work provides an empirical foundation for trustworthy decision-making in emergency response contexts.

Abstract

During disasters, extracting causal relations from social media can strengthen situational awareness by identifying factors linked to casualties, physical damage, infrastructure disruption, and cascading impacts. However, disaster-related posts are often informal, fragmented, and context-dependent, and they may describe personal experiences rather than explicit causal relations. In this work, we examine whether Large Language Models (LLMs) can effectively extract causal relations from disaster-related social media posts. To this end, we (1) propose an expert-grounded evaluation framework that compares LLM-generated causal graphs with reference graphs derived from disaster-specific reports and (2) assess whether the extracted relations are supported by post-event evidence or instead reflect model priors. Our findings highlight both the potential and risks of using LLMs for causal relation extraction in disaster decision-support systems.

Problem

• causal relation extraction
• social media
• disaster intelligence
• Large Language Models
• situational awareness

Innovation

• Large Language Models
• Causal Relation Extraction
• Disaster Intelligence
• Validation Framework
• Social Media Analysis