TrustMap: Mapping Truthfulness Stance of Social Media Posts on Factual Claims for Geographical Analysis

2025-04-09

AI Summary

This paper addresses the problem of identifying public stances toward factual claims on social media and analyzing their geographic distribution. To this end, we propose TrustMap—a novel framework that jointly models stance detection and fine-grained geographic analysis for the first time. TrustMap employs retrieval-augmented fine-tuning of language models to achieve high-accuracy three-way stance classification (support, oppose, neutral), and integrates location-aware aggregation with interactive geospatial mapping to enable both claim-level and region-level exploration of factual belief patterns. Its key contribution lies in uncovering state-level stance polarization and uncertainty hotspots across diverse factual claims in the United States. Evaluated on real-world social media data, TrustMap demonstrates superior stance classification accuracy and provides actionable insights into the spatial heterogeneity of factual beliefs.

Technology Category

• Natural Language Processing: Fact-Checking / Misinformation Detection (NLP Focus)
• Knowledge Representation and Reasoning: Reasoning with Beliefs
• Data Mining & Knowledge Management: Representing, Reasoning, and Using Provenance, Trust

Application Category

• Social Networks and Social Media: Media and governance, opinion dynamics, filter bubbles, polarization
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Web Mining and Content Analysis: Sentiment analysis and opinion mining

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.

Problem

• Mapping public stances on factual claims geographically
• Classifying social media posts by truthfulness stance
• Analyzing regional variations in public engagement with claims

Innovation

• Retrieval-augmented model for stance classification
• Fine-tuned language models for automatic analysis
• Geographical grouping of social media stance patterns