Understanding User Profiles on Social Media for Fake News Detection

TLDR

Social media’s rapid, inexpensive dissemination has made it a popular news source, yet its lower quality compared to traditional outlets fuels widespread fake news, making detection crucial and prompting interest in using user engagement data to improve accuracy. The study aims to understand how user profiles on social media correlate with fake news by building datasets that distinguish experienced from naïve users. We construct real‑world datasets measuring users’ trust levels on fake news and identify experienced versus naïve user groups. Comparative analysis of explicit and implicit profile features shows that experienced and naïve users can be distinguished, providing a foundation for future automatic fake news detection.

Abstract

Consuming news from social media is becoming increasingly popular nowadays. Social media brings benefits to users due to the inherent nature of fast dissemination, cheap cost, and easy access. However, the quality of news is considered lower than traditional news outlets, resulting in large amounts of fake news. Detecting fake news becomes very important and is attracting increasing attention due to the detrimental effects on individuals and the society. The performance of detecting fake news only from content is generally not satisfactory, and it is suggested to incorporate user social engagements as auxiliary information to improve fake news detection. Thus it necessitates an in-depth understanding of the correlation between user profiles on social media and fake news. In this paper, we construct real-world datasets measuring users trust level on fake news and select representative groups of both "experienced" users who are able to recognize fake news items as false and "naïve" users who are more likely to believe fake news. We perform a comparative analysis over explicit and implicit profile features between these user groups, which reveals their potential to differentiate fake news. The findings of this paper lay the foundation for future automatic fake news detection research.