Bridging the Urban Divide: Adaptive Cross-City Learning for Disaster Sentiment Understanding

2026-02-15

AI Summary

This study addresses the pervasive urban-centric bias in existing social media–based disaster sentiment analysis, which often overlooks marginalized and vulnerable communities. To mitigate this limitation, the authors propose an adaptive cross-city learning framework that integrates human mobility signals with city similarity–driven data augmentation to enable joint behavioral-textual modeling and multimodal fusion. This approach represents the first effort to explicitly link mobility patterns with sentiment expression, uncovering geographically diverse affective dynamics during disasters. Evaluated on the 2025 Southern California wildfires, the model achieves state-of-the-art performance by accurately capturing delayed emotional responses and sentiment shifts in overlapping impact zones, while also demonstrating a statistically significant positive correlation between observed mobility behavior and expressed sentiment.

Technology Category

• Natural Language Processing: Fact-Checking / Misinformation Detection (NLP Focus)
• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Search and Optimization: Learning to Search

Application Category

• Web Mining and Content Analysis: Sentiment analysis and opinion mining
• Social Networks and Social Media: Fairness and bias in social network and social media analysis
• Search and Retrieval-Augmented AI: Web query analysis, representation and understanding

Abstract

Social media platforms provide a real-time lens into public sentiment during natural disasters; however, models built solely on textual data often reinforce urban-centric biases and overlook underrepresented communities. This paper introduces an adaptive cross-city learning framework that enhances disaster sentiment understanding by integrating mobility-informed behavioral signals and city similarity-based data augmentation. Focusing on the January 2025 Southern California wildfires, our model achieves state-of-the-art performance and reveals geographically diverse sentiment patterns, particularly in areas experiencing overlapping fire exposure or delayed emergency responses. We further identify positive correlations between emotional expressions and real-world mobility shifts, underscoring the value of combining behavioral and textual features. Through extensive experiments, we demonstrate that multimodal fusion and city-aware training significantly improve both accuracy and fairness. Collectively, these findings highlight the importance of context-sensitive sentiment modeling and provide actionable insights toward developing more inclusive and equitable disaster response systems.

Problem

• urban bias
• disaster sentiment
• underrepresented communities
• social media
• sentiment understanding

Innovation

• adaptive cross-city learning
• mobility-informed behavioral signals
• city similarity-based data augmentation
• multimodal fusion
• fair disaster sentiment analysis