TrustMap: Mapping Truthfulness Stance of Social Media Posts on Factual Claims for Geographical Analysis
Zhengyuan Zhu, Haiqi Zhang, Zeyu Zhang, Chengkai Li
TL;DR
TrustMap addresses the need to map public perceptions of factual claims across regions by integrating retrieval-augmented truthfulness stance detection with geospatial visualization. It builds a pipeline over Politifact claims and claim–tweet pairs from X, applying the RATSD model to classify stances (positive, negative, neutral) and visualize distributions by state and city over time. The work contributes the first public map of truthfulness stances across topics, a scalable stance-detection framework, and an interactive UI for topic- and region-specific analyses, supported by comparative performance evaluations. These findings illuminate how communities engage with truth and misinformation and offer a tool to support targeted fact-checking and public awareness, while acknowledging data-access constraints, geolocation noise, and ethical considerations.
Abstract
Factual claims and misinformation circulate widely on social media and affect how people form opinions and make decisions. This paper presents a truthfulness stance map (TrustMap), an application that identifies and maps public stances toward factual claims across U.S. regions. Each social media post is classified as positive, negative, or neutral/no stance, based on whether it believes a factual claim is true or false, expresses uncertainty about the truthfulness, or does not explicitly take a position on the claim's truthfulness. The tool uses a retrieval-augmented model with fine-tuned language models for automatic stance classification. The stance classification results and social media posts are grouped by location to show how stance patterns vary geographically. TrustMap allows users to explore these patterns by claim and region and connects stance detection with geographical analysis to better understand public engagement with factual claims.