Easy-access online social media metrics can foster the identification of misinformation sharing users

2024-08-27

AI Summary

This study addresses the early identification of high-risk users in social media misinformation propagation. We propose a low-barrier detection method leveraging publicly available, easily obtainable indicators—such as daily tweet volume, account age, and follower count—on X (formerly Twitter). Through statistical analysis and regression modeling, we systematically demonstrate, for the first time, the predictive power of these simple network metrics for content veracity (i.e., propensity to disseminate misinformation). Results show that posting frequency exhibits a significant positive association with misinformation dissemination, account age shows a negative association, and follower count acts as a critical moderating variable. The approach requires no private user data or computationally intensive models, ensuring high scalability and practical deployability. It provides a lightweight, interpretable, and operationally feasible tool for large-scale misinformation source tracing and timely intervention.

Technology Category

• Natural Language Processing: Fact-Checking / Misinformation Detection (NLP Focus)
• Application Domains: Misinformation & Fake News
• Data Mining & Knowledge Management: Representing, Reasoning, and Using Provenance, Trust

Application Category

• Social Networks and Social Media: Influence propagation, information diffusion, and the prediction on networks
• Web Mining and Content Analysis: Content-based information diffusion
• Responsible Web: Mitigating misinformation and disinformation

Abstract

Misinformation poses a significant challenge studied extensively by researchers, yet acquiring data to identify primary sharers is time-consuming and challenging. To address this, we propose a low-barrier approach to differentiate social media users who are more likely to share misinformation from those who are less likely. Leveraging insights from previous studies, we demonstrate that easy-access online social network metrics -- average daily tweet count, and account age -- can be leveraged to help identify potential low factuality content spreaders on X (previously known as Twitter). We find that higher tweet frequency is positively associated with low factuality in shared content, while account age is negatively associated with it. We also find that some of the effects, namely the effect of the number of accounts followed and the number of tweets produced, differ depending on the number of followers a user has. Our findings show that relying on these easy-access social network metrics could serve as a low-barrier approach for initial identification of users who are more likely to spread misinformation, and therefore contribute to combating misinformation effectively on social media platforms.

Problem

• Identifying misinformation sharers using accessible social media metrics
• Differentiating users by likelihood of sharing low factuality content
• Leveraging tweet frequency and account age to combat misinformation

Innovation

• Uses average daily tweet count
• Leverages account age metrics
• Analyzes follower-dependent effects