Authors: JooYoung Lee, Emily Booth, Hany Farid, Marian-Andrei Rizoiu

Published on: March 13, 2024

Impact Score: 8.0

Arxiv code: Arxiv:2403.08391

Summary

• What is new: This research shifts the perspective on online misinformation, arguing that it’s not just about the factual accuracy of the content shared but also about the content selection and presentation styles that echo users’ biases.
• Why this is important: The challenge of understanding why misinformation spreads online, specifically among communities with fringe ideologies.
• What the research proposes: Investigation into the role of consensus-based content, topic selection by far-right users, and differences in writing styles between moderate and far-right news publishers in spreading misinformation.
• Results: Far-right and moderate news publishers shared similar information quality, but far-right users selectively shared topics aligning with their biases. Differences in writing style between moderate and far-right publishers were significant, and users likely to share misinformation could be identified by their communication style.

Technical Details

Technological frameworks used: Content analysis, machine learning for style detection

Models used: Natural Language Processing (NLP) techniques

Data used: Australian news content, Twitter data

Potential Impact

Social media platforms, news organizations, companies in the digital trust and safety space

Want to implement this idea in a business?

We have generated a startup concept here: InfoTune.