Abstract:
Social media's role in news consumption is a two-edged sword. On one hand, its accessibility, affordability, and swift information sharing encourage people to turn to social media for news. However, on the other hand, it also allows for the rampant spread of "fake news" – news with intentionally false information. The proliferation of fake news poses significant risks to individuals and society. Consequently, the detection of fake news on social media has become a prominent research area, garnering substantial attention.
Detecting fake news on social media indeed poses distinct challenges and exhibits unique characteristics that make traditional detection algorithms inadequate or unsuitable for this task. Firstly, fake news is deliberately crafted to deceive readers with false information, making its detection based solely on news content difficult and complex. Hence, auxiliary information, such as users' social engagements on social media, becomes crucial in making accurate determinations. Secondly, leveraging this auxiliary information poses its own set of challenges, as users' interactions with fake news generate large, incomplete, unstructured, and noisy data.
Given the challenging and relevant nature of fake news detection on social media, to facilitate additional investigation into this issue, we ran a survey. The survey provides a comprehensive review of detecting fake news on social media, covering various aspects such as fake news characteristics based on psychology and social theories, existing algorithms from a data mining perspective, evaluation metrics, and representative datasets. We also discuss related fields of study, unresolved issues, and future research paths for identifying bogus news on social media.
